Escherichia coli is a commonly used chassis microorganism in synthetic biology because there is a wide range of information available regarding its characteristics and well-defined experimental protocols.
Our ultimate aim was to improve the understanding of one of the lactic acid bacteria required in yogurt production: Streptococcus thermophilus. S. thermophilus could be incredibly useful in synthetic biology, specifically in creating yogurt with increased production of different macromolecules – this is an application where E. coli would not be suitable.
Although there are other bacteria present in yogurt (most importantly, Lactobacillus delbrueckii subsp. bulgaricus), we chose to use S. thermophilus because it is a naturally competent organism: certain strains of S. thermophilus are highly transformable – for example, the strain we used (LMD-9) has been shown to yield up to 106 transformants per ml of culture [1]. Competence can also be induced through adding a 24-amino-acid hydrophobic peptide (ComS) in strains that are not as easily transformed [2] This takes advantage of the ComRS system, whereby ComS associates with the Rgg-like regulator ComR in order to induce the transcription of comX. ComX encodes the sigma factor σX, which upregulates genes which are required for DNA transformation Cite error: Invalid <ref> tag;
refs with no content must have a name. A protocol for transformation using ComS can be found on our protocols page.
Shuttle Vector
E. coli is an exceptionally easy to work with organism due to how thoroughly it has been researched – therefore, we planned to do all our BioBrick assembly in E. coli, and then transform S. thermophilus with the constructs. To easily transfer these constructs between E. coli and S. thermophilus, a BioBrick compatible shuttle vector between these two organisms had to be constructed. We utilised pMG36ET, a shuttle vector already used in previous S. thermophilus research [3]. Key features of this plasmid include a gene encoding erythromycin resistance and a pWV01 origin of replication, which has a broad host range. Due to the presence of the pWV01 origin of replication, the shuttle vector does have a very low copy number in comparison to pSB1C3. The plasmid had to be made BioBrick compatible by cutting at the highlighted EcoRI and PstI sites and inserting an oligo with the BioBrick prefix and suffix (Figure 1).
Figure 1: Snapgene plasmid maps of a) empty pMG36ET with BioBrick restriction sites highlighted b) pMG36ET with J04450 BioBrick construct inserted between existing EcoRI and PstI restriction sites
We were able to transform both E. coli and S. thermophilus with this plasmid, resulting in erythromycin-resistant colonies. By using this plasmid as a shuttle vector, we were able to clone BioBrick constructs in E. coli, and then purify the DNA for transformation of S. thermophilus. During the course of the project, we successfully transformed S. thermophilus with an amilCP BioBrick construct in this shuttle vector - therefore, we have confirmed that the plasmid is suitable for use in both E. coli and S. thermophilus.
Quantification of Promoter Strength
Three strong constitutive promoters have already been used in S. thermophilus. These are P32, P25 [4], and phlbA [5]. We wanted to check the strength of these promoters in E. coli to see whether it is viable to make a single BioBrick construct for use in both E. coli and S. thermophilus. During the course of the project, we confirmed that all three promoters are suitable for use with E. coli: a high level of fluorescence was produced by GFP when ligated downstream of each of the three S. thermophilus native promoters (Figures 2&3). Fluorescence measurements were also taken for GFP under the expression of multiple Anderson promoters (J23100, J23101, J23105, J23106, J23113 and J23117). As these promoters are so well characterised, these measurements were used as a benchmark to compare the native S. thermophilus promoters to.
In order to take these measurements, each promoter was ligated to a GFP-coding region which was attached to either a B0032 or B0034 ribosome binding site. Controls were also set up: I13500 and E5501 were used to check the level of GFP produced under no promoter, and non-transformed cells were used as a negative control. The constructs were then each transformed into TOP10 E. coli, and three biological replicates were cultured overnight. A plate reader was then used to measure the GFP fluorescence of the overnight cultures. Three technical replicates were carried out for each overnight culture, giving a total of 9 replicates per BioBrick construct.
A graph was constructed with the data collected showing the fluorescence of GFP under expression of the various promoters and ribosome binding sites B0034 (on the right) and B0032 (on the left). A plate reader was used which measured the OD of every sample simultaneously to the fluorescence measurements, so OD was normalised across all samples in order to get comparable fluorescence readings.
Figure 2: Graph of strength of fluorescence expressed by E. coli cells under 8 different promoters. Measurements with strong RBS B0034 are in light green, and measurements with weak RBS are in a darker green.
Figure 2 shows that the S. thermophiluspromoters were relatively strong, yielding 93% (pHLBA), 81% (p32) and 30% (p25) of the fluorescence values of J23100, the strongest modified Anderson promoter. Further data on this can be found on our measurements page. Fluorescence levels in the constructs using weak RBS B0032 were significantly reduced, but relative promoter strengths were equivalent. The unusually high level of fluorescence for promoter p25 with B0032 (Figure 3) can be attributed to the fact that the plasmid had dimerised. The dimerisation was verified with a gel electrophoresis but due to time constraints we were unable to carry out a second set of fluorescence measurements in order to obtain a correct value for fluorescence of GFP under p25.
Our observations show that the three S. thermophilus native promoters have significant variances in strength. Therefore, we have found a satisfactory range of promoters for expression experiments in S. thermophilus. Due to the different strengths, these promoters can be utilised for a wide range of purposes in synthetic biology research.
AmilCP
Although GFP is already available as a useful reporter gene in S. thermophilus, we wanted to explore alternatives. AmilCP is a blue chromoprotein which was first submitted to the BioBrick registry ([BBa_K592009]) by the Uppsala team in 2011. It produces a strong blue pigment within 24 hours of incubation when transformed into E. coli. The gene is useful as a reporter due to the fact it can be seen very easily by the naked eye. Therefore, it allows for qualitative analyses more quickly than GFP. We wanted to see whether we could use it in S. thermophilus, as it has previously been characterised in E. coli, but not in our bacteria.
Figure 4: K592025 BioBrick ligated to p25 in psb1C3 and transformed into E. coli. A vibrant blue colour can be seen in transformant colonies.
We created a promoter-RBS-gene construct in our shuttle vector and transformed this into E. coli to ensure that blue pigment was still produced. Due to the low copy number of the plasmid, the colonies were less pigmented than the ones pictured above; however, it could still be seen from the plate that amilCP was being produced. After extracting the DNA and transforming it into S. thermophilus, the resultant colonies remained white.
We verified the presence of amilCP through carrying out a miniprep of S. thermophilus and checking that a plasmid of the correct size was present on an agarose gel. After confirming this, TOP10E. coli was transformed with the S. thermophilus miniprep. The transformant E. coli cells were pale blue in colour, showing the production of amilCP was happening in both bacteria.
Figure 5: Pale blue TOP10 E. coli cells which have been transformed with an S. thermophilus miniprep containing the shuttle vector plasmid with amilCP. The bottom row of cells are control cells producing no pigment.
Despite the presence of amilCP in the transformed S. thermophilus, no pigment was produced even after several days. We have come up with three hypotheses as to why this could have occurred:
- The ribosome binding site B0034 was not suitable for S. thermophilus- therefore, translation of the gene was not able to occur and blue chromoprotein was not synthesised. This is unlikely due to the consensus sequences of ribosome binding sites for both organisms being similar, with the only nucleotide base difference between the two being energetically equivalent (ACCTCCTTT/A).
- Each of the BioBrick parts in the construct was suitable for S. thermophilus, but the cellular conditions meant that the expressed amilCP was degraded before any visible pigmentation could develop.
- The low copy number of the plasmid meant that although amilCP was expressed, it was not expressed in a large enough amount to be visible to the naked eye.
We cannot directly conclude that amilCP is not a suitable reporter for S. thermophilus. Further testing will be required to ensure that the ribosome binding site, promoters, and vector used were appropriate for S. thermophilus, before deciding that amilCP is not useful in this organism.
