Difference between revisions of "Team:iTesla-SoundBio/Project/results"

Synthesis of a Biobrick-compatible Factor C Plasmid in E. coli

Figure 1.1

Click to see Legend

Well 1, empty; Well 2, ladder; Well 3, Fragment 1 PCR; Well 4, Fragment 2 PCR. Fragment 2 did not amplify during PCR. The Fragment 1 PCR yielded DNA at around 1.5 kb, the desired length, but a nonspecific band of DNA around 0.95 kb also showed up.

Figure 1.2

Click to see Legend

Well 1, empty; Well 2, ladder; Well 3, Fragment 1 PCR; Well 4, Fragment Well 1, ladder; Well 2, empty; Well 3, Fragment 1 PCR; Well 4, empty; Well 5, Fragment 1 PCR. There was amplification, but the DNA that appeared in the gel was less than 1 kb, which is much shorter than the expected length (~ 1.5 kb). There were also smears.

One of our biggest issues was that few PCRs of our Factor C fragments were successful. Results often showed nonspecific bands,smears or bands that were shorter than the desired length (as shown in Figures 1.1 and 1.2). We realized late in the year that our biobricking primers were poorly designed and did not have enough bases matching our fragments, so they were annealing incorrectly. We tried re-designing primers and adjusting the annealing temperature and extension times, but we still could not get the PCRs working efficiently. We still attempted to purify some PCRs and use them in cloning, but the poor result of our PCRs was one of the biggest factors in our failed transformations which are discussed below.

Figure 2.1

Click to see Legend

Row 1, no DNA negative control; Row 2, positive control; Row 3, No insert; Row 4, experimental (should have had factor C fragment, no mRFP). The red colonies on the no insert plate indicate that there was digested plasmid that ligated back togetherand the mRFP was not removed and replaced with Factor C successfully.

Figure 2.2

Click to see Legend

Top: No insert control; Bottom: experimental.

Figure 2.3

Click to see Legend

Digest of BBa_J04450 (mRFP) + pSB1C3. The upper bands (pSB1C3) were gel extracted and used for ligation with the factor C fragments.

Our transformations were consistently unsuccessful. We kept having the same issue: the presence of red colonies. The gel above is an example of the results from our restriction digests of BBa_J04450 with PstI and EcoRI; they consistently worked and the mRFP was cut out of the pSB1C3 vector, giving 2 distinct bands. We attempted gel extractions of pSB1C3 (upper band on figure # 2.3, ~ 2 kb), but our bacteria still expressed mRFP after ligation and transformation. There were some white colonies, but we believe they were two vectors ligated together and did not contain our desired factor C fragment. Since we continued having this issue, we attempted to transform E. coli using an alternative method called FastCloning [1].

Figure 3.1

Figure 3.2

In the FastClone method, the vector and insert are designed to have overlapping regions at the site of insertion. The biobrick prefix and suffix can serve as this overlap. First, the vector (pSB1C3) and insert (Factor C fragment) are amplified using PCR with Phusion or another high fidelity polymerase. During PCR, Phusion can exhibit 3’ exonuclease activity that creates sticky ends at each 3’ end of the amplified strand of DNA. The PCR reactions of the vector and insert are then mixed together and digested with DpnI, a restriction enzyme that only cuts methylated DNA; in other words, DpnI will cut the template plasmid DNA (provided it was derived from an organism that methylates DNA like E. coli) used in PCR to prevent it from being expressed by bacteria. The overlapping regions allow the insert and vector to anneal when mixed and form a nicked plasmid which is then ligated in vivo.

Before working with pAX01 or Factor C, we attempted a practice transformation using the FastClone method. We attempted to insert a CFP gene into the pSB1C3 backbone, but as seen in Figure 3 above, none of the colonies that grew on the plates expressed CFP, and some expressed mRFP as in the earlier transformations. We also had fewer transformants. We suspect that the DpnI digest did not destroy all of the template DNA, and the only DNA that the bacteria were transformed with was the original BBa_J04450 plasmid or two pSB1C3 vectors that annealed together. The purpose of shifting our focus to FastCloning was to quickly and easily clone the full Factor C gene into E. coli, but it was ultimately unsuccessful.

Synthesis of BioBrick-compatible pAX01

We also hoped to make our pAX01 plasmid biobrick compatible. There is currently a version of the pAX01 plasmid in the iGEM registry; however, our goal was to submit a version of the plasmid that also contained the xylose-inducible expression system. We designed several pairs of primers for site-directed mutagenesis that would remove all of the illegal cut sites. The proposed process designed by our team is a series of PCRs and transformations that removes and inserts new pieces until we attain our desired biobrick, but there was difficulty in primer design and we had little success with our PCR experiments. There were ultimately no successful transformations, and we were unable to create the biobrick part of pAX01 that we designed.

Synthesis of Factor C-producing B. subtilis

In parallel to the attempted biobricking work, we worked towards transforming B. subtilis 168 with our chosen integration vector, pAX01. Integration with pAX01 would be a necessary part of expressing Factor C, so it was important to confirm that it worked properly even though we were unable to successfully clone Factor C. Upon successful integration of pAX01 into the genome, B. subtilis will become resistant to erythromycin. Based on this antibiotic selection, we had some apparently successful transformations, albeit with low efficiency and after many attempts.

Figure 4.1

Click to see Legend

Single successful transformation of Bacillus subtilis 168 with pAX01.

Figure 4.2

Click to see Legend

Plate streaked using the single transformant depicted in Figure 4.1. The left side of the plate is regular Bacillus subtilis (no growth), and the right is the transformation.

Figure 5.1

Click to see Legend

Wells 1 & 3, empty; Well 3, ladder; Wells 4-9, colony PCRs. This figure shows a gel of several colony PCRs, none of which worked. There was no amplification, and the only DNA present was genomic DNA that stayed in the wells.

Figure 5.2

Click to see Legend

Well 1, ladder; Wells 2-4, B. subtilis, primer pair 1; Wells 5-7, Transformed B. subtilis, primer pair 1; Wells 8-10 B. subtilis, primer pair 2; Wells 11-13, Transformed B. subtilis, primer pair 2; Well 14, pAX01, primer pair 1; Well 15, pAX01, primer pair 2.
The figure above is a gel of several colony PCRs of regular B. subtilis and B. subtilis transformed with pAX01. None of the colony PCRs on our transformants amplified any DNA as seen in this gel. Furthermore, of the PCRs that did work, only one gave a single band of the expected size (Well 8), but others gave nonspecific bands. pAX01 should not have amplified at all with primer pair 1, but there was still a band of DNA (well 14).

We then attempted a colony PCR on our transformants as a preliminary method of confirming successful integration. We designed primers to amplify the integrated portion of DNA between portions of the B. subtilis 168 genome. Our colony PCRs gave us mixed results. Sometimes the PCR experiments resulted in no apparent amplification and there was only genomic DNA stuck in the wells (figure 5.1). In other attempts, one primer pair gave us the desired results but also showed nonspecific bands from other parts of the B. subtilis genome (figure 5.2). Since many colony PCRs of our transformations did not work at all, we predicted that what we thought was transformed B. subtilis was not in fact B. subtilis and was instead contamination from another organism. The colonies of the transformations grew unusually compared to our regular B. subtilis, and we had several issues with contamination of buffers when attempting transformations. So, in addition to doing colony PCRs designed to confirm successful integration, we also amplified the 16s region of our bacteria so that we could then sequence it and confirm the organism’s identity.

After sequencing the 16s region of B. subtilis and our transformation and running a BLAST on the data, results showed that our transformed bacteria were very likely B. subtilis and are definitely a Bacillus species. Although we cannot confirm whether or not it is the desired strain (B. subtilis 168) with these data, considering we did not work with any other B. subtilis strains, it is unlikely that it is something else.

The fact that it is likely B. subtilis 168, not other bacteria that had contaminated our transformations, is a promising sign that the transformations actually worked as expected, and pAX01 was able to successfully integrate into the genome.

Interestingly, there were slight differences between our stock B. subtilis and the transformants in the 16s regions. There were 3-4 single nucleotide insertions or deletions. As these single nucleotide polymorphisms occurred near the beginning and ends of the identified sequences, where Sanger sequencing tends to be the least accurate, we believe they were minor sequencing inaccuracies. Further attempts at sequencing could better help us understand if these were real mutations or errors in the sequencing process. Despite these possible mutations, our transformations are still identifiable as B. subtilis, and pAX01 is very likely still able to work just as effectively. Working integration is an important step towards eventually cloning the Factor C gene and producing the protein in B. subtilis.

We have also redesigned primers for confirming successful integration into the genome through colony PCR as our earlier results with previous primers were insufficient. After colony PCR, we hope to obtain concrete sequencing data that shows successful integration of pAX01. Upon successful integration, pAX01 will insert a sequence of interest into the lacA locus in the genome. Therefore, in the sequencing results, we expect to see a portion of the genome adjacent to the lacA gene (found using data from NCBI) next to a portion of our inserted sequence, including the xylose cassette and the erm region (erythromycin resistance selection marker). We expect to have concrete sequencing data soon that shows that the pAX01 integration vector worked as expected.

References