Results
Our summer of hard work resulted in a successful assembly of the biobrick BBa_K2474000. Several experiments were made to validate our biobrick. First we sent our constructed plasmid to sequencing. The result from this showed us a 99 % match on those 77 % that the sequencing covered (40 % from VF2 and 37 % from VR) which confirms that we succeeded with the assembly.
For validating the function of the biobrick we choose to search for amyloid-beta with antibodies and look for fluorescence from mNeonGreen and how this was regulated with the promoter. For this we used a dot-blot and spectrophotometric analysis of fluorescence.
Dot-blot
Read more about the dot-blot here
The results from the dot-blot can be seen in figure 1 compared to figure 2. Table 1 is a scheme explaining figure 1.
From these figures, we could confirm that our bacteria was expressing proteins which have epitopes that the antibodies, specific to amyloid-beta, could bind to. This is confirmed by comparing the 4 marks in the second column with the 4 negative marks in the first. By the black marks in figure 1, where we added cell lysate from induced bl21 containing our biobrick, we confirm the expression of amyloid-beta. From the weaker marks we confirm there is some leakage from the promoter.
Table 1. scheme over cell lysate applied to the filter.
| Empty Bl21 |
uninduced Bl21 with BBa_K2474000 |
| Empty Bl21 |
uninduced Bl21 with BBa_K2474000 |
| Empty Bl21 |
induced Bl21 with BBa_K2474000 |
| Empty Bl21 |
induced Bl21 with BBa_K2474000 |
To the left: Figure 1. Results from dot-blot. To the right: Figure 2. Filter appearance viewed with the naked eye.
Spectrophotometric analysis
E.coli BL21 bacteria cultures with BBa_K2474000 was prepared overnight and the OD was later set to 0.4. The bacteria was distributed to a 96-well plate so that each well contained 200 µl. These were then induced by arabinose to express the fluorescent fusion protein Amyloid-Beta mNeonGreen. The fluorescence was then analysed on a plate reader. The original intent of this experiment was to analyse protein expression over time. However due to technical difficulties this was not possible. Instead figure 3 belowrepresents emission spectra of the fluorescence from Amyloid beta mNeongreen 24 hours after induction.
Figure 3. They were distributed in a 96 well plate in 8 replicate at each of the 4 induction levels (5 mM, 1 mM, 0.2 mM and 0 mM). Empty bacteria (BL21(DE3)) were also included on the plate to serve as a control. All measurements were done with an infinite M1000 pro plate reader. Due to technical difficulties and lack of an mNeonGreen fluorescent standard curve it was impossible to present this data in the preferred unit (µM/OD). The data points represent the mean value of the replicates and the error bars represent 95 % confidence intervals.
From this graph we can clearly see that the promoter and fluorescent parts of our biobrick works as expected as the fluorescence is highly proportional to the arabinose concentration. We can also see that the promoter have a small leakage in its regulation as uninduced bacteria with BBa_K2474000 (purple in figure 3) show more fluorescence than empty bacteria.
This graph in combination with the dot-blot above (figure 1) confirms that our biobrick BBa_K2474000 is fully functional.
Expression optimization
Sadly we never got to test the expression of our fusion proteins in combination with the chaperones. This is due to the fact that we had trouble with the assembling of the plasmids. We eventually succeeded with BBa_K2474000 but then we had no more time. The exact problem with our assembly process isn’t something we have experimentally confirmed, but we have some ideas. Probably it’s due to a compatibility flaw with DNA-secondary-structure and the assembly method we used. The methods we used for assembling plasmids involves 5’ exonuclease to create ssDNA and then the method involves connecting two complementary ssDNA segments [1]. Therefore we designed fragments with exactly the same end-sequence, overhangs, to make them match after reacting with the exonuclease.
What we didn’t consider when we started to design our DNA-fragments was eventual secondary structure in the DNA. The problem probably arose when we created overhangs between our fragments and the vectors pSB1C3 and pSB1A3. Here we included the terminating sequence from the vectors. These terminating sequences are stem loops, a form of secondary structure occurring in ssDNA. This means that the ssDNA-strands, created by the exonuclease, would react with itself and bind strongly. This would result in the complementary sequences being locked in the stem loops making joining them impossible. This is what we think happened with all our assemblies except the last one.
The Future of the project would be to redesign the fragments to make them compatible with our chosen method of assembling. This could potentially be done by shortening the matching overhangs on the fragments to not contain the stem loop structure. Hopefully this could solve the problem, but another method could be to create artificial overhangs on the psb1C3 with primers and use these instead. With the assembled plasmids the laboratory work could continue experimenting with conditions and testing of the model.
[1] https://www.neb.com/-/media/catalog/datacards-or-manuals/manuale5520.pdf
