Team:KU Leuven/Project/Glucosemodel/EBF

In this part, we will give more information about the E-β-farnesene (EBF) part of the project. EBF is an alarm pheromone, released by almost all of the 4000 aphid species known thus far in response to the presence of predators (e.g. the ladybug) or other disturbances. In response to the produced EBF, aphids change their metabolism and turn into a winged form, allowing them to "flee the scene" and thus increase their survival rate. Apart from the short term repelling effect, EBF can also cause long term effects: changes in aphid’s development, fecundity, survival when introduced to different growth stages, etc. These have a negative impact on aphid reproduction. Natural aphid predators such as the ladybugs are also attracted by EBF.

We cloned and expressed the EBF synthase gene in E. coli. This enzyme will break down (2E,6E)-farnesyl diphosphate into (E)-β-farnesene (EBF) and diphosphate (see reaction scheme below).
The enzyme prefers bivalent cations as cofactors; a Mg²⁺ concentration of 5 mM should be beneficial for EBF synthase function. The ideal pH for EBF synthase will be between 5.5 and 7.

The Model and the Genes

The EBF construct we designed consists of a constitutive promoter with a lac operator, the EBF synthase itself and a double terminator. We used BBa_B0015 for the double terminator. EBF is not only made by aphids but also by plants and other organisms in a form of biomimicry. We obtained two different sources of the EBF gene. One gene originates from the soil bacterium Streptomyces coelicolor (Centre of Microbial and Plant Genetics of KU Leuven). We chose this plant-residing bacterium because it would be a perfect chassis for the ultimate expression of EBF in our E. coligy system. The other EBF gene is from the plant Artemisia annua (sweet wormwood) and was a kind gift from Professor Peter Brodelius (Kalmar University, Sweden). Here we were inspired with the plant origin. The K_M for the Artemisia annua protein is calculated at 0.0021 mM, with a Kcat/K_M=4.5 and a turnover number of 0.0095 s^-1. For the Streptomyces coelicolor protein the K_M is 0.0168 mM and the turnover number 0.019 s^-1.
Unfortunately, the EBF synthase from Streptomyces coelicolor is a bifunctional enzyme, not only processing β-farnesene but also containing albaflavenone synthase activity. For this reason, we chose to follow up on the Artemisia annua gene and product. For our construct, our first choice was a medium strength promoter with medium RBS (BBa_K608006); we nonetheless also made the construct with a strong promoter and RBS. The lac operator in front of the EBF synthase gene will allow us to switch the transcription of the EBF synthase gene on and off.

The pathway to E-β-Farnesene

β-farnesene is a terpenoid that is converted from farnesyl pyrophosphate (FPP) by the enzyme β-farnesene synthase (EC 4.2.3.47).
FPP is the precursor of β-farnesene, that is produced by the building blocks, the molecules isopentenyl pyrophosphate (IPP) and its isomer dimethylallylpyrophosphate (DMAPP).
These precursors of farnesyl pyrophosphate can be produced by several metabolic pathways. Most prokaryotes use the non-mevalonate or DXP pathway, producing IPP starting from glyceraldehyde-3-phosphate and pyruvate. Eukaryotes, except for plants, exclusively use the mevalonate pathway, producing IPP starting from acetyl-CoA. Plants use both pathways.

Mevalonate pathway, showing the conversion of acetyl-CoA to general terpenoid precursor IPP and its isomer DMAPP.

On the left you can see the non-mevalonate pathway or DXP pathway, showing the conversion of pyruvate and glyceraldehyde-3-phosphate to the terpenoid precursor IPP and its isomer DMAPP.
Pyr = pyruvate, G3P = glyceraldehyde-3-phosphate, DXP = 1-deoxy-D-xylulose 5-phosphate, MEP = 2-C-methylerythritol 4-phosphate, CDP-ME = 4-phosphocytidyl-2-C-methylerythritol, CDP-MEP = 4-phosphocytidyl-2-C-methyl-D-erythritol 2-phosphate, MEcPP = 2-C-methyl-D-erythritol 2,4-cyclopyrophosphate, HMB-PP = (E)-4-Hydroxy-3-methyl-but-2-enyl pyrophosphate, DXS = DXP synthase, DXR = DXP reductase, CMS = CDP-ME synthase, CMK = CDP-ME kinase, MCS = MEcPP synthase, HDS = HMB-PP synthase, HDR = HMB-PP reductase

Problems & Solutions Concerning The Pathway

FPP is an important precursor, used for the biosynthesis of lots and lots of compounds. Once we insert a plasmid containing the β-farnesene synthase gene, we may obtain only a very small amount of β-farnesene, since the precursor amount wasn't increased and there simply isn’t enough FPP available to produce the amount of β-farnesene to fully use the capacity of the EBF synthase enzyme we brought in.
A solution may be to co-transform plasmids to engineer a mevalonate pathway in E. coli, thereby upregulating the production of FPP. This larger amount of FPP may then be converted to β-farnesene, creating a large enough amount of this volatile. This was demonstrated many times by J.D. Keasling in S. cerervisiae, while Martin et al., (2003) implemented this mevalonate pathway in E. coli. In the article, they described their successful efforts to create a high level production of amorphadiene by introducing the mevalonate pathway in E. coli. However, expression of this heterologous pathway led to such an abundance of isoprenoid precursors that cells ceased to grow or mutated to overcome the toxicity. This once again shows the need for a controlled production of the elements in this pathway; too much is equally detrimental as too little.

The genes involved in the mevalonate pathway
(Martin et al.)

Since there are eight genes responsible for the mevalonate pathway, Martin et al. decided to split them up into two parts. A first plasmid named pMevT, responsible for the conversion of acetyl-CoA to mevalonate, harboring the atoB, HMGS and tHMGR genes into a pBAD33 vector, and a second one named pMBIS, harboring the ERG12, ERG8, MVD1, idi and ispA genes into a pBBR1MCS-3 plasmid. Coexpression of these two operons in an ispC deficient E. coli strain produced the terpenes, even in the absence of mevalonate, indicating that the mevalonate pathway works. A combined expression of their recombinant mevalonate pathway and the synthetic gene product (ADS in their case) resulted in greatly improved yields.

Even though we do not need a very high production of EBF it would be definitely better to optimise the pathway, by using the plasmids pMevT and pMBIS, described above. Implementing them into our BanAphids along with the synthetic β-farnesene synthase gene could result into high yields of β-farnesene. This way the amount of EBF can be easily changed via the amount of bacteria used or the concentration of the cofactor Mg²⁺.
Due to the short amount of time iGEM offered we did not yet started doing this, but this is definitely something future teams might look into.

To confirm that our E. coli strain containing our EBF synthase producing brick BBa_K1060009 produced EBF, we turned to GC-MS analysis with the help of Dr. Jan Baeten (Centre for Surface Chemistry and Catalysis; Prof. Dirk De Vos). Supernatant of bacterial cultures were extacted 3 times with hexane. GC-MS analyses were carried out using a 7890A Agilent gas chromatograph coupled to a 5977A mass spectrometric detector. The GC was equiped with a HP-5MS capillary column (30m x 0.25 mm x 0.25 µm). 1 µl of each sample was injected using splitless (head pressure 9.15 psi) or pulsed splitless (head pressure 20 psi) injection at a temperature of 250 °C. The initial oven temperature of 40 °C was held for 1 min., ramped at 6 °C / min. to 124 °C, then ramped at 20 °C / min. to 320 °C and finally kept at this temperature for 5 min. The transfer line and ion source were held at 320 °C and 230 °C, respectively. Mass spectra were taken between masses m/z 30-300 with an ionization potential of 70 eV. Retention indices of standards were determined by co-injection of a C7-C30 n-alkane mixture (Supelco) and were compared with published retention indices. To test if EBF could be detected by this setup, farnesene (Sigma-Aldrich, W383902, mixture of isomers) was used as a standard. In total 16-17 peaks were observed of which the following could be identified with the help of the mass spec and retention index (RI). E-beta-farnesene RI = 1459 (lit. 1458-59), (3Z,6E)-alfa-farnesene RI = 1490 (lit. 1487) and (3E,6E)-alfa-farnesene RI = 1506 (lit. 1506-1510). All samples were measured in duplicate; once in SCAN-modus, splitless injectie of 1 µl, and once in SIM-modus (most sensitive), pulsed splitless injectie of 4 µl. No EBF could be detected for our samples, suggesting that the concentration of EBF produced in our cultures is below the threshold values.

Gettin' the gene

In the case of the EBF synthase gene from Streptomyces coelicolor, we amplified this gene with a colony PCR. The EBF synthase gene from Artemisia annua was received in the pET28 vector from professor Brodelius (Kalmar University, Sweden). In this gene an additional EcoRI restriction site was present, which would conflict with the standard iGEM cloning work. Therefore we removed this site via site directed mutagenesis after transferring the gene into the iGEM pSB1C3 backbone.

Cutting and pasting

Once we obtained the target gene (EBF) in the standard pSB1C3 backbone, we started our cloning work. We used plasmid pSB1C3 with a promoter or terminator as chassis, cut this open and inserted the gene of interest. When ligating the insert in front of the double terminator, we cut the vector with EcoRI and XbaI, and the insert with EcoRI and SpeI. The promotor vector on the other hand is cut with SpeI and PstI restriction sites, and the insert is cut with XbaI and PstI restriction sites. This works because SpeI and XbaI are isoschizomers.
Ligations were performed in parallel in two different ways. In one setup we ligated for 20 minutes at 16 ℃, and in comparison, the second ligation of the same products was conducted at 16 ℃ overnight.
For transformation, we used both chemically competent cells and electrocompetent cells. Electroporation had a higher efficiency when compared to heat shock transformation.

Confirmation

After we observed colonies the next day, we needed to confirm the products. The first step we did was usually a colony PCR to check if the insert was in the vector, this was followed up by digestion confirmation after the plasmid extraction. Only the plasmids which succeeded in both controls were send for sequencing, the final confirmation.

gBlocks

Meanwhile, we also built the EBF construct with a lac operator between the promoter and gene, using the gBlock principle. We designed the gBlocks, assembled them and ligated the insert into pSB1C3 backbone. The colonies obtained also went through the three confirmation steps mentioned above before we were satisfied.

For more details on the labwork and the wetlab difficulties as well as how we overcame them, please consult our wetlab journal.

Kajiwara S., Fraser P., Kondo K., Misawa N., Expression of an exogenous isopentenyl diphosphate isomerase gene enhances isoprenoid synthesis in Escherichia coli, Biochem J. 324, 421-426 (1997).
Martin V., Pitera D., Withers S., Newman J., Keasling J., Engineering a mevalonate pathway in Escherichia coli for production of terpenoids, Nature Biotechnology 21(7), 796-802 (2003).