Results and Discussion
The control constructs in pETDuet-1 vector, described in experimental design, were ordered at the end of June to start performing the experiments during the summer. Unfortunately, several logistical problems and other unforeseen issues arose and severely delayed the intended schedule. In addition, full access to lab space and necessary equipment were not provided until the beginning of September. Due to the delays, the constructs were received in mid September and throughout October, resulting in limited time to run the planned experiments. Priorities had to be made and the focus shifted to assembling BioBricks. Further, a backup plan was executed through modifying the constructs to suit BioBrick standards and order them as gBlocks. Several attempts to assemble these gBlocks, which arrived prior to the pETDuet-1 constructs, into BioBricks were performed. However, assembling and cloning successfully takes a great amount of time which we were not fortunate to have.
One frequently tested part was GFP1-9, see parts. Cells were transformed with the ligation product and the resulting colonies were cultured overnight. After cell harvest, plasmids carrying the insert were isolated and agarose gel electrophoresis was run. The result is displayed in fig. 1 with two distinct bands visualized in the corresponding well. The band with a size of 1.6 kbp is believed to be supercoiled and the other band (with a size of >3 kb) is thought to be nicked plasmid. The other two wells are displaying PCR products of the plasmid carrying GFP1-9 mentioned earlier. A band corresponding to a size around 300 bp is most likely the result of an empty plasmid since that is the length of the sequence between the primers. A plasmid containing the gene of interest would have a size of approximately 900 bp. Furthermore, the obtained smearing could be due to wrong annealing temperature, which was further investigated using a temperature gradient. It could also be the result of using too many amplification cycles, however, this does not explain why the DNA ladder also appears as smeared. Similar results were obtained with the other parts as well.
Various sources of error were encountered on the way and had to be investigated and handled, as described in the upcoming section. Due to these complications and restricted time to troubleshoot and redo the necessary experiments, we are sad to say that no BioBrick samples could be submitted to the registry.
All cells used in the experiments were in-house made competent cells, which could have affected the competency of the cells. However, after each batch, cell competency was tested showing a sufficient transformation efficiency.
Possible sources of error during the digestion could be due to too few additional nucleotides at the end of synthesized gBlocks. This could have lead to limited performance of restriction enzymes. Hence, it could be theorized that no thorough cleavage could be performed, resulting in poor production of sticky ends. During the first experiments, the digestion time could have been too short. In the following experiments the digestion time was extended, however, no improvements were obtained. Another possibility of poor digestion could be due to addition of an insufficient amount of DNA.
At first, the ligation time was considered too short and was later extended to an overnight ligation in order to improve the outcome. It is also possible that the optimal ratio of insert to backbone was not used, although three different ratios were tested each time. Problems with the ligation could also be a result of poor digestion.
The resulting low transformation yields could be due to low competency of cells, non-gentle handling of cells during the transformation process or a result of unsuccessful ligation. Although several bacterial strains and DNA concentrations were tested, the result remained the same.
We did extensive troubleshooting and changed the protocols multiple times to no avail. The changes that were carried out were all discussed with faculty members to ensure relevancy. With more time, or possibly as recommendations for future work, we would of course be interested in testing the constructs according to our experimental design. As we never got the opportunity to try the complete circuit, it is not possible to say whether this design would have worked or not. The original construct comprised of genes encoding three different proteins could pose a risk of putting too much pressure on the cell when inducing protein expression. Instead, an option could be attempting double transformation using compatible plasmids. Another interesting experiment would be to purify the proteins from the biosensor device to see how they interact ex situ. Finally, it would of course be desirable to test the biosensor using natural ligands - organic pollutants and phthalates.