Plasmids used during the characterization of our biobrick
For simplicity, our biobrick BBa_K2671420 is called GroES in the rest of the text. The Takara-plasmid pGroE7 is used to exprees both GroES and GroEL. This system is often referred to as GroE. We chose to use GroE to better understand the GroES/GroEL-system and to investigate the importance of overexpression of GroE. It is important to note that there is always a GroE expression in the bacteria. In all experiments, there is a presence of GroE although it is lower when GroE is not overexpressed with pGroE7.
The red segment of the plasmids represents the antibiotic resistance for each plasmid. The green segment represents the promoters and the yellow segment represents the origin of replication (ORI).
As seen in Figure 1, we had to ligate our biobrick into pSB4A5 to avoid having the same ORI and the same antibiotic resistance as in any of the other plasmids we planned to use. We also chose the tetracycline promoter to be enable to fine-tune the expression of our biobrick.
Figure 1 illustrates plasmids used in our experiments. pSub*, can either be mNG-Aß1-42, EGFP-Aß1-42, α-synuclein-EGFP or Tau0N4R-EGFP. They were used as a client-protein for our co-expressions.
A simple explanation on how we achieved our experimental results is shown in Figure 2. Four combinations were introduced into E.coli (BL21). Substrate (A) , substrate and our biobrick (A+C), substrate, our biobrick and GroE (A+B+C), substrate and GroE (A+C).
A can be either mNG-Aß1-42, EGFP-Aß1-42, α-synuclein-EGFP and Tau0N4R-EGFP. B is always pGroE7 and C is pSB4A5-GroES.
Bacteria containing plasmids were grown in LB- media containing antibiotic for respective plasmid until desired OD was achieved. The chaperones were induced with 0.25 mg/ml L-arabinose and 200 ng/ml tetracycline. For the substrate plasmids we used 0.5 mM IPTG, and the substrate was induced 30 mins after the chaperone plasmids. The bacterial samples were placed in a 96-well plate that ran for 16 hours in 37 °C in a Galaxy FLUOstar microplate reader. Measurements were conducted every 15 minutes.
Figure 2. Illustrates the experimental procedure for our co-expression. The plasmids used are: Substrate plasmid (A), pGroE7 from Takara (B) and GroES plasmid (C). The substrate plasmids contain one of the following client proteins: mNG-Aß1-42, EGFP-Aß1-42, α-synuclein-EGFP and Tau0N4R-EGFP.
Results from the experimental measurements are shown in Figure 3. The y-axis represents the fluorescence intensity at 520 nm. The excitation was carried out at 485 nm. The x-axis represents time in hours. The second set of graphs shown for each substrate represent the normalized values for the fluorescence intensity. This was analyzed because it represents the kinetic growth of the substrate proteins.
As seen in the graphs, 50% of the maximum intensity has been marked with a dotted line. This makes it easy to interpret if the folding of the different substrate proteins was affected by the chaperone systems. To be able to interpret the normalized values, no error bars are shown.
Results for the mNG-Aß1-42 substrate protein
As seen in the top three graphs, our biobrick increases the intensity of the mNG-Aß1-42 protein. However, an interesting notice is that in concert with the pGroE7 plasmid, the results create the impression that the folding rate slows significantly, as shown in the bottom graphs.
A high concentration of GroES, with or without GroEL seems to be the best fit for this substrate protein. A double-expressed GroES-gene together with GroEL seems to prevent the production of the substrate more than helping it.
Figure 3. Results for mNG-Aß1-42. Top graphs show fluorescene intensity divided by the start OD600 over 16 hours at 37°C. The bottom graphs show the normalized values from the top graphs. The error bars represents the Standard Deviation.
Results for the EGFP-Aß1-42 substrate protein
The results for EGFP-Aß1-42 co-expression can be seen in Figure 4. The protein looks quite different compared to the mNG-Aß1-42 protein. In this case, our biobrick and the other two combinations show an increase in fluorescence intensity compared to the substrate alone. An interesting notice from the bottom three graphs is that our biobrick without the presence of GroEL slows down the folding rate, and a decrease in intensity is not shown after 16 hours.
Figure 4. Results for EGFP-Aß1-42. Top graphs show fluorescene intensity divided by the start OD600 over 16 hours at 37°C. The bottom graphs show the normalized values from the top graphs. The error bars represents the Standard Deviation.
Results for the α-synuclein-EGFP substrate protein
The results for the co-expression of α-synuclein-EGFP (Figure 5) shows almost no difference in intensity which can be seen in the top left graph. However, in concert with GroEL it shows a clear increase. Same thing goes for co-expressing GroES with GroE, though it increases more slowly compared with expression of only the substrate.
Notably as shown in the three bottom graphs, the kinetics of α-synuclein-EGFP is slowed down a lot by the combination with our biobrick(GroES) and pGroE7(GroE). This is not seen in the bacteria with only the substrate and pGroE7.
Figure 5. Results for a-synuclein-EGFP. Top graphs show fluorescene intensity divided by the start OD600 over 16 hours at 37°C. The bottom graphs show the normalized values from the top graphs. The error bars represents the Standard Deviation.
Results for the Tau0N4R-EGFP substrate protein
The last co-expression was with Tau0N4R-EGFP (Figure 6). The top graphs without our biobrick (the blue lines) show a clear increase in intensity, and a small decrease in the folding rate as seen in the bottom graph. Interesting to note is that the bacteria with our biobrick(GroES) and the pGroE7 (GroE) slowed down the folding rate by a large margin. It is also clear that there is no decrease in the intensity at any point in this graph with this combination. The same results can be applied for the plasmid with only GroE.
Figure 6. Results for Tau0N4R-EGFP. Top graphs show fluorescene intensity divided by the start OD600 over 16 hours at 37°C. The bottom graphs show the normalized values from the top graphs. The error bars represents the Standard Deviation
mNG-Aß1-42: As previously mentioned, the folding rate slows down when pGroE7 is present. The combination with a GroES and GroE seems to prevent the production of the substrate. An explanation for this could be that mNG-Aß1-42 is fairly good at folding itself and that the presence of chaperones may instead inhibit the folding.
The best fit for the substrate protein is our biobrick, or a high concentration of GroE, which means that our biobrick and GroE-plasmid is in this case best for increasing the amount of substrate and give the highest folding rate.
As seen in Figure 4, when adding chaperones to the systems, the concentration of substrate increases compared to without adding GroES or GroE. The graph presenting the results from our biobrick has not reached its maximized concentration of the substrate. Potential future studies is needed in order to find out the maximized concentration of the substrate. This could be done with an increased experimental time.
The graphs presenting the folding rate of each expressed system shows little difference.
When expressing α-synuclein-EGFP with GroES there is little difference in concentration compared to expressing α-synuclein-EGFP without a chaperone plasmid. With GroE the concentration of the substrate greatly increases compared to all other expression systems. The kinetics of the folding rate for α-synuclein-EGFP expressed with GroE and GroES is noticeably decreased. It seems that when GroEL is expressed with an increased concentration of GroES, it decreases the folding rate of this substrate. If GroEL is not expressed though, there is no difference in the folding rate.
There is no decrease in intensity for Tau0N4R-EGFP combined with chaperones and a study with longer experiment time can be useful to study the maximum concentration. The folding rate of all substrates with chaperones slows down, but with difference in how much. In presence with GroES the folding rate is higher than with GroES and GroE co-expressed. With GroE, the folding rate is in between the others.
Conclusion and Future
The purpose of our project was to improve the yield of recombinant protein when expressed in bacteria, by utilizing a co-expression system with chaperones. The data indicates that our part works as expected and helps the folding of aggregation prone proteins. As suspected, not all proteins need this specific help. Seen in 3/4 client proteins were assisted by the co-expression of GroES. However, one was not. We suspect GroES has a holdase activity if it interacts on its own. It would be interesting to compare it to an known holdase chaperone. It would also be interesting to investigate the binding mechanism between GroES and substrate, through perhaps crystallization or HSQC.