"
Team:St Andrews/Omega-3-synthesis
From 2012.igem.org
Omega-3 fatty acid synthesis
Introduction
ω-3 Fatty acids are an essential component of our diet and are paramount to maintaining human health. Our team is recreating this synthetic pathway in E. coli, using genes from the cyanobacteria Synechocystis and the trypanosomatid Leishmania major. Combining the DNA code for elongase and desaturase enzymes, we can convert the plain fatty acid of E. coli into highly valuable ω-3 fatty acids.
Project Description
-
Fi. 1
"Synechocystis sp." -
Fi. 2
"Trypanosome cruzi" -
Fi. 3
"Leishmania major"
Omega-3 fatty acids are an essential part of the human diet (Bender, Bender, 1999). Human beings, as all larger organisms, cannot synthesize ω-3 fatty acids, This is due to a lack of the enzyme Δ15 desaturase, which creates a double bond at the 15th carbon of a long-chain fatty acid. Certain microrganisms, such as microalgae and cyanobacteria, do contain this desaturase and can thus directly synthesize ω-3 fatty acids (Arts et al, 2009). ω-3 fatty acids then enter the food chain – algae are eaten by fish, and seafood is sbsequently the main source of ω-3 for humans(Tonon et al, 2002).
However, the current economic policies of overfishing are a serious contributor to marine biodestruction. As the human population is estimated to rise to 9.1 billion by 2050 (Cohen, 2003), pressure on fish stock will increase. Additionally, global warming will reduce the availability of ω-3 (Arts et al, 2009): at higher temperatures, microalgae produce less ω-3 desaturated fatty acids. Desaturated carbon chains cause a lower melting temperature in the membrane, which the microorganism wants to avoid by synthesizing more saturated fatty acids in their membranes (Garwin, Cronan, 1980). Thus, the combination of declining fish stock and a decrease in overall ω-3 fatty acids is making the supply for human nutrition a relevant issue.
Harvesting algae directly is costly and ineffective (Borowitzka, 1997). There is much potential in expressing a metabolic pathway for ω-3 fatty acid synthesis in E. coli, which is cheaper and more accessible.
Synthesizing the pathway
Fi. 4
"The metabolic pathway to ω-3 fatty acids"modified from Livore et al, 2006
E. coli naturally synthesize poly-unsaturated fatty acids up to a carbon chain length of 18, with a single desaturation at the 9th carbon (18:1) (Marr, Ingraham, 1969). Valuable ω-3 fatty acids require 3 double bonds, and need to have a chain length between 18 and 22 carbons. The double bonds needs to start at the thrid carbon, counting from the end of the carbon chain.
In order to have
The genes for Δ12, Δ15 (ω6) and Δ6 were obtained from Synechocystis sp., a cyanobacterium. The trypanosomatid Leishmania major provided the DNA for the ELO 6 gene. Additionally, we used Trypanosome cruzi as a secondary source of Δ12. (cf. Fig. 2)
However, our first successful ligations of Δ12 did not yield us with the expected 18:2 fatty acid. We hypothesized that E. coli’s inherent 18-carbon chain fatty acid might not be suited as a substrate for Δ12 – perhaps the double bond is in a different position. Therefore, we "fed" our cells with suitable 18:1, to then observe 18:2 fatty acid in the mass spec results!
Methods
Sequences
By combining a number of genes, we were able to recreate the pathway described above (Fig. 4). The following genes were employed (please click for sequences and KEGG numbers):
PCR
These genes were extracted using PCR (Promega, GoTaq HotStart) at temperatures 48°C and 56°C. Both vectors pET-15b and pET-20b were tested for expression. The recombinant plasmids were ligated into two E. coli strains: BL21(DE3) and BL21(DE3)pLysS.
Fig. 5: "UV photograph of PCR results"
The figure shows the results of a PCR extraction of our genes of choice, done with GoTaq HotStart PCR kit at 2 different annealing temperatures: Δ12 (48°C) - Δ12 (56°C) - Δ15 (48°C) - Δ15 (56°C) - Δ6 (48°C) - Δ6 (56°C).
Primers
All primers are notated 5' to 3'. Initially, we worked with NdeI and XhoI as the restriction sites.
When using the pET-duet vector, we needed additionally primers for the alternative restriction sites and His-tags. For Δ6 and Δ12, we cut with HindIII and EcoRI; NdeI and XhoI were used for Δ15 and ELO6.
After a number of expression attempts, some initial conclusions were reached. All experiments done on ELO6 failed. Also, Δ12 from T. cruzi gave overall weaker results than the same gene from Synechocystis. As such, latter work was only carried out on genes from Synechocystis.
Δ6, Δ12, and Δ15 was successfully expressed in pET-15b and transformed into cell strains BL21(DE3) and BL21(DE3)PlysS. However, functionality could not be established. It was hypothesized that the naturally-occuring 18:1 fatty acid in E. coli is the wrong substrate for the desaturases. This fatty acid has its desaturation at the 11th carbon. Thus, the E. coli were manually fed 18:1 (Δ9).
Additionally, a pET-duet vector expressing Δ12 and Δ15 was used to start assembling the metabolic pathway. Due to time constraints, no duet vector BioBrickTM was submitted.
Analysis
Lipids were extracted from transformed E. coli using the Blight/Dyer method. (Please refer to the Lipid extraction lab book entry for further detail.)
Lipids were extracted from transformed E. coli using the Blight/Dyer method. (Please refer to the Lipid extraction lab book entry for further detail.)
We then analyzed the the fatty acid content of the transformed cells to determine whether it differed from the previously analyzed background lipid composition. This was done using mass spectrometry.
A Bradford protein assay was carried out to asses protein concentration (Bradford, 1976).
We sent off samples to GATC for sequencing.
Results
Bradford protein assay
Chart 1: "Bradford protein assay"
This chart shows the results of a Bradford protein assay. It was carried out on the following ligated plasmids transformed into BL21: Δ12 in pET-15b (in 2 colonies), Δ12 in pLYS, Δ12 in pET-duet (in 2 colonies), Δ6 in pET-duet, Δ12/Δ15 in pET-duet (in 2 colonies).
The Bradford protein assay discuss results
Mass Spectrometry
-
Chart 2: "Fatty acid methyl ester analysis"
C18:3 refers to an 18 carbon backbone with 3 double bonds. C18:2 and C18:1 would have 2 double bonds and 1 respectively. The upper portion of this figure shows the peaks that would be expected for the various fatty acid methyl esters.The lower 4 panels show mass spec fingerprints that are indicative of our fatty acids of interest.
-
Chart 3: "Lipid extraction analysis"
All of the cellular lipids were extracted and their fatty acid methyl esters produced. These were analysed by mass spec. In the control cells no C18:2 is present. When Δ6 is expressed, a peak indicative of C18:2 is present. However, the primary fatty acid is still C18:1 - with the bouble bond present at position 9. With the expression of Δ12 and the addition of exogenous C18:1(3) we get great abundance of C18:2 and negligible abundance of C18:1 indicating that the C18:1 is being processed successfully. The lowest graph shows that Δ12 expressed with Δ15 acts in a complementary fashion and C18:3 - Omega 3 - is produced.
-
Chart 4: "Membrane assay"
The cellular membranes were extracted and their lipids were analysed by mass spec. As in chart 2 it is confirmed that expression of Δ12 and Δ15 lead to the production of omega 3 fatty acids. This chart shows that these lipids locate to the plasma membrane of the bacteria
Sequencing results
- 16/7/12 Δ15
- Δ15 and Δ6 desaturases from Synechocystis in pET-15b were sent off for sequencing. Once again, the results were discouraging even though protein and lipid analysis showed the opposite. BLAST showed that the gene found in both samples corresponded to a T. brucei HMG-CoA reductase which we thought we had cut out from the plasmid. We thus performed a PCR, using the vectors sent off for sequencing as a template.
- 16/7/12 Δ6
- 29/6/12 ELO6
- After showing successful insertion of L. major Δ6 elongase in pET-15b, the sample was sent off for sequencing. Unfortunately, the gene sequenced was not the Δ6 elongase we were hoping to find, but a T. brucei mevalonate kinase, which was previously inserted in the pET-15b vectors we had been using. Thus, we digested some fresh pET-15b to be used in the future.
- 23/7/12 Δ12
- 15b forward gave 0 nucleotides
- 15b reverse gave 0 nucleotides
- 20b forward gave a 41 nucleotide result.
- A BLAST search showed this to be compatible with Crocosphaera watsonii (genome shotgun sequence, 21%), a diazotrophic cyanobacteria.
- 20b reverse gave 0 nucleotides
BiobricksTM
TODO
References
ARTS, et al, 2009. Lipids in Aquatic Ecosystems. New York. Springer.
BENDER D. A. and BENDER, A. E, 1999. Benders' dictionary of nutrition and food technology. Cambridge: CRC Press.
BRADFORD, M., 1976. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem, 7;72:248-54.
COHEN, J., 2003. Human Population: The Next Half Century. Science, New Series, Vol. 302, No. 5648. Pp. 1172-1175.
GARWIN J. L. and CRONAN J. E. Jr, 1980. Thermal modulation of fatty acid synthesis in Escherichia coli does not involve de novo enzyme synthesis. J Bacteriol, 141(3): 1457–1459.
LIVORE V., TRIPODI K., UTARRO A., 2007. Elongation of polyunsaturated fatty acids in trypanosomatids. FEBS Journal, 274: 264–274.
MARR A., INGRAHAM J., 1969. Effect of temperature on the composition of fatty acids in Escherichia coli. J. Bactiol. 84(6). Pp. 1260–1267.
TONON T., et al, 2002. Long chain polyunsaturated fatty acid production and partitioning to triacylglycerols in four microalgae. Phytochemistry, Vol 61 Iss 1. Pgs 15-24.
