Team:Uppsala/Zea-Strain
ZEAXANTHIN-STRAIN
In our project we chose to concentrate on the pathway that leads from farnesyl pyrophosphate (FPP) to crocin. The whole pathway consists of eight genes that code for eight enzymes which might make integrating all of the genes into a plasmid and keeping the plasmid in the bacteria more difficult. Dividing the pathway and integrating the part that leads from FPP to zeaxanthin into the chromosome would both give us a stable zeaxanthin-producing Escherichia Coli strain and make performing the remaining steps easier.
Resulting Zeaxanthin Producing E. coli Strain
Our team has successfully integrated the whole zeaxanthin pathway into the chromosome using lambda red recombineering (figure 1) and extracted zeaxanthin (figure 3). This will hopefully give other iGEM teams more freedom to work with and build on carotenoid pathways and make zeaxanthin more affordable to use in experiments.
We also combined this strain with the BioBrick that contains the crocin pathway. To see the combined result, click over to the Result page. In the future it would be good to integrate the whole pathway from farnesyl pyrophosphate (FPP) to crocin into the chromosome for a stable crocin producing strain that does not require antibiotic selection which would make it easier to use as for example food coloring.
How we did it
Zeaxanthin has previously been expressed in E. coli by iGEM teams using a plasmid. We decided to integrate this pathway into the chromosome using the Lambda red recombineering method. This would give our project several advantages such as releasing all the plasmid origins and cassettes which would make the insertion of the crocin pathway genes or any other genes of a pathway that originates from zeaxanthin easier. It would make the strain more stable because no constant selective pressure is needed and makes it possible to introduce larger constructs and longer pathways. This also means that there is no need for the use of antibiotics which makes the purification process easier, especially if the product is later used for nutritional purposes. And of course since the first step of our zeaxanthin pathway – farnesyl pyrophosphate – is endogenous to E. coli we would be able to express the whole pathway from farnesyl pyrophosphate to crocin with no need for costly intermediates.
If we use cat-sacB selection/counterselection in Lambda red recombineering (which we did) we get a scarless method that does not leave behind resistance markers. Lambda red is based on homologous recombination which is mediated by bacteriophage lambda proteins and usually requires only about 35 base pairs of homology on both sides of the inserted gene to work.
The pathway that leads from FPP to zeaxanthin (figure 2) includes five genes: crtE, crtB, crtI, crtY and crtZ. All these genes have been previously BioBricked into an operon for E. coli. Due to the time limitations of the project we decided to synthesize two constructs each containing two of the genes – crtEB and crtZY. For successful integration with Lambda red recombination the inserts should not exceed the size of 3000 bp and that is why we could not synthesize all of the genes as a single construct. Each construct also contained promoter, ribosome-binding sites and the necessary homologies for the future integration into the strains. The crtI gene was amplified from BioBrick (iGEM Slovenia 2010) with the primers that contained the homologies.
The E. coli strains that were used for recombination contained a pSIM5-tet temperature sensitive plasmid with the lambda red system and tetracycline resistance. When temperature rises above 37 °C the lambda red enzymes are expressed. They also had cat-sacB selection cassette (based on Uppsalas own BioBrick BBa_K864150) carrying chloramphenicol resistance gene and Bacillus subtilis levansucrase sacB gene that is lethal for gram-negative bacteria when expressed in presence of sucrose.
Three different starting strains were used during the experiments. We also used three different promoters to avoid homology among these regions. The strain genotypes were:
1. Eco ∆gsp::cat-sacB /pSIM5-tet
2. Eco ∆bglGFB::cat-sacB /pSIM6
3. Eco ∆IS150::CP25-cat-sacB /pSIM5-tet
Lambda red recombineering was performed in three steps. For more detailed information, please, see the protocol for Lambda red and the Lab Notebook.
Step 1
Phytoene dehydrogenase gene crtI was amplified with PCR from zeaxanthin BioBrick and the homologies were added with the primers. Lambda red was performed to replace cat-sacB-T0 selection cassette with crtI, and the cells were plated on sucrose agar plates to select for transformants that had successfully integrated the crtI gene.
Eco ∆IS150::CP25-cat-sacB-T0 /pSIM5-tet + crtI (PCR product) → Lambda Red (sucR) → Eco ∆IS150::CP25-crtI /pSIM5-tet
A construct with beta-carotene hydroxylase gene crtZ and lycopene cyclase gene crtY was inserted with lambda red into another E. coli strain.
Eco ∆gsp::cat-sacB /pSIM5-tet + crtZY (PCR product) → Lambda Red (sucR) → Eco ∆gsp::pFAB70-crtZY /pSIM5-tet
Step 2
The trimethoprim resistance gene dhfr (dihydrofolate reductase) was inserted with lambda red into the strain that now contained crtZY genes. The genes we placed very close to each other to enable the following transduction with P1 bacteriophage.
Eco ∆gsp::pFAB70-crtZY /pSIM5-tet + dhfr (PCR product) → Lambda Red (tmpR) → Eco ∆gsp::pFAB70-crtZY gspC::dhfr
Transduction with P1 bacteriophage was done to transfer cat-sacB-T0 selection cassette from a donor strain into the strain containing crtI gene.
Eco ∆IS150::CP25-crtI /pSIM5-tet + ∆bglGFB::cat-sacB (P1 lysate) → Transduction (camR) → Eco ∆IS150::CP25-crtI ∆bgl::cat-sacB /pSIM5-tet
Step 3
Lambda red was performed to replace this newly inserted cat-sacB-T0 cassette with a construct containing geranylgeranyl diphosphate synthase gene crtE and phytoene synthase gene crtB. The clones with the correct sequence changed color to light red. Our lycopene producing strain was constructed!
Eco ∆IS150::CP25-crtI ∆bgl::cat-sacB /pSIM5-tet + pFAB46-crtEB (PCR product) → Lambda Red (sucR) → Eco ∆IS150::CP25-crtI ∆bgl::pFAB46-crtEB → Lycopene!
Another transduction with P1 bacteriophage was done with the strain containing crtZY as a donor strain and the strain containing crtEBI as a recipient strain. The clones with the correct sequence turned bright yellow. We got our zeaxanthin producing strain with the whole pathway from FPP to zeaxanthin integrated into the chromosome!
Eco ∆IS150::CP25-crtI ∆bgl::pFAB46-crtEB + ∆gsp::pFAB70-crtZY gspC::dhfr (P1 lysate) → Transduction (tmpR) → Eco IS150::CP25-crtI ∆bgl::pFAB46-crtEB ∆gsp::pFAB70-crtZY gspC::dhfr → Zeaxanthin!
All of the steps were confirmed with PCR, gel electrophoresis and sequencing. We were able to extract and purify the expensive yellow zeaxanthin compound from our strain, see figure 3.
Besides observing the extracted zeaxanthin by eye we performed an absorbance measurement in the UV-Vis spectra (figure 4). Here we compared the absorbance spectra after extraction from the zeaxanthin strain, from wildtype E. coli and a zeaxanthin standard. The zeaxanthin strain had two peaks at 460 and 482 nm which were not present in wildtype E. coli. These peaks were also present in the standard, therefore we can conclude that our produced strain produces zeaxanthin. For the measurements the extracted compounds were dissolved in toluene.
Our team would like to express a special appreciation and thanks to Erik Wistrand-Yuen who provided us with the starting strains and the protocols for the lambda red method and who spent hours guiding and instructing us and providing practical help. Without him this success would not have been possible.
