From 2012.igem.org

Limonene

Responsible: Lara Kuntz and Andrea Richter

A Common Saying Is "If Life Gives You Lemons - Make Lemonade". For Us It Is Rather "If Life Gives You Limonene - Make Beer".

The obvious thing to do, isn't it? Beer with lemonade is a very popular beverage throughout Germany, e.g. "Radler", "Alsterwasser", "Russ'n". So why not take the shortcut and skip the intermediary?

Just think of the refreshing sensation of lemons paired with the complex richness of a chilled brew.

Doesn't that make you thirsty? Then you're probably an Englishman -

Just think of the refreshing sensation of lemons paired with the complex richness of a chilled tepid brew.

d-limonene is used as a component of flavorings and fragrances since it has an orange/lemon-like odor. Limonene has been shown to inhibit rat mammary and other tumor development http://www.ncbi.nlm.nih.gov/pubmed?term=Tsuda limonene 2004 Tsuda et al., 2004. Being an excellent solvent of cholesterol, d-limonene also has been used clinically to dissolve cholesterol-containing gallstones. Because of its gastric acid neutralizing effect and its support of normal peristalsis, it has also been used for relief of heartburn http://www.ncbi.nlm.nih.gov/pubmed/18072821 Sun, 2007.

Producing the flavoring substance limonene in our beer might result in a fresh, lemon-like taste on the one hand. On the other hand, we might have beneficial effects on health such as preventive activity against cancer, dissolution of gallstones and relief of heartburn.

We are close to achieving our goal of a limonene tasting beer. We have successfully cloned (+)-Limonene synthase 1 into our newly established yeast expression vector pTUM100. We transformed Saccharomyces cerevisiae with the plasmid carrying limonene synthase. We proved expression of limonene synthase via western blot. We verified the functionality of (+)-limonene synthase 1 by in vitro assay with purified limonene synthase on the one hand. On the other hand, we proved functional limonene production in the yeast culture via headspace GC-MS. Furthermore we analyzed differences in protein expression in yeast depending on existence of the yeast consensus sequence. To survey the toxic concentration of limonene for yeast cells, a toxicity assay was performed. We showed an inhibition of growth at 1 mM and a lethal effect at 100 mM. These concentrations should not be reached by expression of limonene synthase in yeast.

Unfortunately we could not yet prove a significant content of limonene in self brewed beer. We will try to integrate the limonene synthase expression cassette into the genome of yeast and afterwards we will repeat the brewing experiment.

Background and Principles

Limonene is a cyclic terpene and a major constituent of several citrus oils (orange, lemon, mandarin, lime and grapefruit). It is a chiral liquid with the molecular mass of 136.24 g/mol. The (R)-enantiomer smells like oranges and is content of many fruits, while the (S)-enantionmer has a piney odor http://www.ncbi.nlm.nih.gov/pubmed/11605760 Fietzek et al., 2001. Therefore, d-limonene ((+)-limonene, (R)-enantiomer) is used as a component of flavorings and fragrances.

Biosynthesis

Limonene is produced by limonene synthase. Limonene synthase uses geranyl pyrophosphate (GPP), which is the universal precursor of monoterpenoids, as educt. (+)-limonene synthase from Citrus limon consists of 606 aminoacids (EC=4.2.3.20) and catalyzes the following reaction: Geranyl pyrophosphate = (+)-(4R)-limonene + diphosphate (see Fig. 1).

Fig. 1: Reaction catalyzed by limonene synthase.

Saccharomyces cerevisiae produces geranyl pyrophosphate via the mevalonate pathway (see Fig. 2). GPP occurs as an intermediate of farnesyl pyrophosphate (FPP) synthesis http://www.ncbi.nlm.nih.gov/pubmed/17096665 Oswald et al., 2007. It has been established that S. cerevisiae has enough free GPP to be used by exogenous monoterpene synthases to produce monoterpenes under laboratory and vinification conditions [[http://www.ncbi.nlm.nih.gov/pubmed/18155949 Herrero et al., 2008], [http://www.ncbi.nlm.nih.gov/pubmed/17096665 Oswald et al., 2007]].

Fig. 2: Simplified isoprenoid pathway in S. cerevisiae, including the branch point to linalool. Abbreviations: HMG-CoA, 3-hydroxy-3-methylglutaryl coenzyme A; IPP, isopentenyl pyrophosphate; GPP, geranyl pyrophosphate; FPP, farnesyl pyrophosphate; DMAPP, dimethylallyl pyrophosphate; HMGR, HMG-CoA reductase; FPPS, FPP synthase; LIS, linalool synthase. [Rico et al., 2010]

Molecular and Physiological Effects of Limonene

Flavor and Aliment

Because of its pleasant citrus flavour and very low toxicity (oral LD50 for mice = 5.6 and 6.6 g/kg body weight), d-limonene is widely used as a flavor and fragrance additive. It is listed in the Code of Federal Regulations as a generally recognized as safe (GRAS) flavoring agent http://www.fda.gov/downloads/Food/FoodIngredientsPackaging/GenerallyRecognizedasSafeGRAS/GRASListings/UCM264589.pdf FDA. It can be found in common food items such as fruit juices, soft drinks, baked goods, ice cream, and pudding in typical concentrations of 50 ppm to 2,500 ppm http://www.ncbi.nlm.nih.gov/pubmed/18072821 Sun, 2007. The normal daily consume of d-limonene is 0.27 mg/kg body weight per day http://www.mri.bund.de/fileadmin/Institute/PBE/Sekundaere_Pflanzenstoffe/Monoterpene.pdf Watzl, 2002. As natural compound of plants, limonene has practical advantages with regard to availability, suitability for oral application, regulatory approval and mechanisms of action. It does not pose a mutagenic, carcinogenic or nephrotoxic risk to humans http://www.ncbi.nlm.nih.gov/pubmed/18072821 Sun, 2007.

Cancer Inhibition

Monoterpenes show anticarcinogenic effects in animal experiments. They have been shown to inhibit rat mammary, gastric, lung and skin tumor development by several discussed mechanisms such as apoptosis induction and modulation of oncogene signal transduction [[http://www.ncbi.nlm.nih.gov/pubmed?term=Tsuda%20limonene%202004 Tsuda et al., 2004], [http://www.mri.bund.de/fileadmin/Institute/PBE/Sekundaere_Pflanzenstoffe/Monoterpene.pdf Watzl, 2002]]. d-limonene induces phase I and phase II carcinogen-metabolizing enzymes such as cytochrome p450. These metabolize carcinogens into less toxic forms on the one hand. On the other hand, they prevent the interaction of chemical carcinogens with DNA. Limonene has also been shown to inhibit tumor cell proliferation, to accelerate the rate of tumor cell death and induce tumor cell differentiation. Furthermore, d-limonene regulates cell growth and/or transformation by inhibiting protein isoprenylation http://www.ncbi.nlm.nih.gov/pubmed/18072821 Sun, 2007.

Solvation of Gallstones

d-limonene is used as excellent solvent of cholesterol, therefore it has been used clinically to dissolve cholesterol-containing gallstones http://www.ncbi.nlm.nih.gov/pubmed/18072821 Sun, 2007. A study with 200 patients showed that direct infusion of 20-30 ml d-limonene (97% solution) completely or partially dissolved gallstones in 141 patients. Gallstones completely dissolved in 96 cases (48%); partial dissolution was observed in 29 cases (14.5%); and in 16 cases (8%) complete dissolution was achieved with the inclusion of hexamethaphosphate (HMP), a chelating agent that can dissolve bilirubin calcium stones http://www.ncbi.nlm.nih.gov/pubmed/1988264 Igimi et al., 1991. Because of its gastric acid neutralizing effect and its support of normal peristalsis, it has also been used for relief of heartburn http://www.ncbi.nlm.nih.gov/pubmed/18072821 Sun, 2007.

Results

BioBricks

Gel Picture of Finished Constructs

Fig. 3: Gel electrophoresis of [http://partsregistry.org/Part:BBa_K801060 K801060] and [http://partsregistry.org/Part:BBa_K801061 K801061] after analytical restrigtion digest with EcoR1 and Pst1.

(+)-Limonene synthase 1 coding region without yeast consensus sequence http://partsregistry.org/Part:BBa_K801061 BBa_K801061 in pSB1C3 is shown next to (+)-limonene synthase 1 with Strep-tag and yeast consensus sequence http://partsregistry.org/Part:BBa_K801060 BBa_K801060 after restriction digest with EcoR1 and Pst1 restriction enzymes. To check success of ligation, DNA fragments were separated by agarose gel-electrophoresis using ethidium bromide as a nucleic acid stain. As expected, the 1665 bp fragment of http://partsregistry.org/Part:BBa_K801061 BBa_K801061 and the 1708 bp fragment of http://partsregistry.org/Part:BBa_K801060 BBa_K801060 were detected additionally to the pSB1C3 vector that has 2070 bp (see Fig. 3).

To compare limonene synthase expression in yeast depending on yeast consensus sequence, we produced duplicates of biobricks; one of each has the consensus sequence, the other one does not. For yeast expression experiments, the biobricks were cloned into the yeast expression vector (pTUM100) recently designed by us.

Investigation of Yeast Consensus Sequence

Fig. 4: Comparison of limonene synthase biobricks ([http://partsregistry.org/Part:BBa_K801065 BBa_K801065] and [http://partsregistry.org/Part:BBa_K801060 BBa_K801060]) with and without yeast consensus sequence.

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC340751/ Hamilton et al., 1987 reported of a consensus sequence upstream of the AUG start codon in yeast. Although not as strong as the mammalian Kozak translation initiation sequence, the yeast consensus sequence is thought to have a 2–3-fold effect on the efficiency of translation initiation http://tools.invitrogen.com/content/sfs/manuals/pyes2_man.pdf pYES2 manual.

We designed duplicates of limonene synthase encoding biobricks; one having the yeast consensus sequence, the other one not having the consensus sequence.

We have only been able to show slight differences in expression between the two biobricks [[http://partsregistry.org/Part:BBa_K801065 BBa_K801065] and [http://partsregistry.org/Part:BBa_K801060 BBa_K801060]] via coomassie staining and western blot. The difficulties of showing the difference via SDS-page may result from variations in the amount of protein applied.

Our in vivo analysis strongly indicates that the consensus sequence does lead to a 2-3-fold enhanced expression in yeast, though (see Fig. 8B). This is consistent with findings of others http://tools.invitrogen.com/content/sfs/manuals/pyes2_man.pdf pYES2 manual.

Purification of Recombinant Limonene Synthase

Fig. 5: Streptavidin Affinity (SA) chromatography.

Fig. 6: Analytical gelfiltration of limonene synthase.

Streptavidin affinity chromatography
The yeast cell extract that was obtained by cell lysis using glass beads of 0.5 mm was centrifuged at 11000 RPM in an SLA-3000 rotor for 60 minutes and subsequently dialysed against 5 liters of 1x Streptavidin Affinity buffer (SA-buffer) over night. The cell extract was then filtrated using a syringe filter with a pore size of 0.45μm and susequently applied on an SA-column. After sample application, the column was washed with 1x SA buffer until a base line was reached. Subsequently, bound protein was eluted using 5 mM Biotin in 1x SA buffer. The chromatogram of the purification is shown on the left side.

Gel Filtration of Purified Protein The protein sample obtained was concentrated using a centrifugation concentrator with a molecular size limit of 30 kDa subsequent filtration. 250μl of the solution were applied to an analytical gel filtration column Superdex 200 10/30 with 1x PBS as running buffer at a flow rate of 0.5 ml/min. In the chromatogram (shown in the figure on the right in section B), there is an aggregate peak at the exclusion limit that may be caused by the preceding concentration and a major peak at an elution volume of 13.580 ml. The calibration line that was obtained from the calibration proteins b-amylase, alcohol dehydrogenase, BSA, ovalbumine, carboanhydrase, cytochrome C and aprotinin filtrated with the same experimental setup resulted in a regression line with the formula y = -39206 x + 3.3463. Using this formula and the elution volume of the limonene synthase, an apparent molecular mass of 70.1 kDa could be determined for the produced limonene synthase. This fits quite well the theoretical molecular mass that was calculated using [http://web.expasy.org/cgi-bin/protparam/protparam ExPASy ProtParam] to be 65977.1 Da. The four kilo daltons difference could be caused by posttranslational modifications. This hypothesis should be tested using mass spectrometry.

In Vitro Detection of Limonene

Fig. 7: Spectrum of in vitro detection of limonene (enzyme assay with [http://partsregistry.org/wiki/index.php?title=Part:BBa_K801060 BBa_K801060]) and reference spectrum.

To test the functionality of purified limonene synthase in vitro, we used an optimized protocol of an enzyme assay with extraction of limonene http://www.ncbi.nlm.nih.gov/pubmed/17662687 Landmann et al, 2007. The limonene synthase was purified via Strep-tag. The enzyme assay was carried out in 25 mM Tris-HCl buffer with 5% Glycerol, 1 mM DTT and cofactors (10 mM MgCl2, 1 mg/ml BSA). 50 µM substrate (geranyl pyrophosphate) and 10 ng purified recombinant limonene synthase were added to the reaction batch. Negative controls were reaction batches without enzyme. The reaction was incubated for 15 min at room temperature. Afterwards, limonene was extracted with pentane, dried with sodiumsulfate and reduced under a stream of nitrogen. Three replicates were done of both sample and negative control.

The pentane extracts were analyzed with gas chromatography-mass spectrometry ("5890 Series II GC" coupled to a "Finnigan Mat 55 S MS") to identify the enzymatically synthesized products.

All enzyme reactions (three replicates) led to the production of limonene while the negative controls did not show limonene. Therefore, we showed that our purified limonene synthase is functional and leads to the production of limonene.

In Vivo Detection of Limonene

Fig. 8: Detection of limonene in headspace above cell culture supernatant. [A] Spectrum of limonene obtained when analyzing cell culture that was transformed with pTUM104 containing construct of [http://partsregistry.org/wiki/index.php?title=Part:BBa_K801060 BBa_K801060]. [B] Overview about different measurements.

Because limonene is a VOC (volatile organic compound) http://www.ncbi.nlm.nih.gov/pubmed/15763095 Pierucci et al., 2005, we expected limonene to be present in the gaseous phase above the cell culture as well as in the cell culture. Therefore, we measured limonene via headspace (SPME needle) GC-MS.

We showed limonene to be produced by yeast that was transformed with pTUM104 carrying limonene synthase coding regions (see [http://partsregistry.org/wiki/index.php?title=Part:BBa_K801061 BBa_K801061] and [http://partsregistry.org/wiki/index.php?title=Part:BBa_K801060 BBa_K801060]).

We detected a greater amount of limonene in the sample that contained limonene synthase with consensus sequence. Hence, we showed that the yeast consensus sequence might increase the expression of limonene synthase and therefore might lead to enhanced limonene production.

Furthermore, we were not able to detect a significant difference between samples that had additional GPP (educt) versus the ones that did not. This might be due to the inability of GPP to diffuse into the cells (hydrophilic character). Since we were able to detect limonene in both samples, it implies that the GPP present in the cells is sufficient for limonene production. This is consistent with the findings of [http://www.ncbi.nlm.nih.gov/pubmed/18155949| [Herrero et al., 2008]] that showed that S. cerevisiae cells (from laboratory and wine strains) contain enough free GPP to be catalytically transformed by monoterpene synthases into monoterpenes.

Detection of Limonene in Beer

Fig. 9: Preparation of sample for GC-MS with SPME.

Fig. 10: iGEM's first and finest SynBio Beer with limonene.

A first attempt to use our genetically engineered yeasts to brew a SynBio Beer were conducted using a transient transfection with a constitutive promoter. The drawback is that in the gyle the selection pressure is not preserved and the loss of the plasmid is possible.

Three liters of gyle were inoculated with 100ml of a stationary yeast culture grown in YPD that was transiently transfected with a plasmid harboring a constitutive expression cassette for the limonene synthase.

We analyzed this first beer for limonene content via headspace (SPME needle) GC-MS. Unfortunately we have not yet been able to proof a significant difference between the beer supposed to contain enhanced amounts of limonene and the negative control beer. This is probably due to a loss of the plasmid which encodes limonene synthase. We will try to integrate the limonene synthase expression cassette into the genome of yeast and afterwards we will repeat the experiment.

Fig. 11: GC spectra obtained from GC-MS analysis of brewed beer. Brown: Beer brewed with yeast that carries pTUM101 with limonene synthase. Green: Beer brewed with yeast that supposedly carries limonene synthase in genome (after genome integration).

Toxicity Assay

To establish whether limonene has an effect on yeast cells, we inoculated three different yeast strains with different concentrations of limonene. Limonene was added to the medium. We examined the laboratory strain INVSc1, a strain which is used for brewing beer and a strain which can be purchased in grocery stores.

At high concentrations, limonene affects the growth of yeast cells. We could show an inhibition of growth at 1 mM and even a lethal effect at 100 mM. At lower concentrations (1 µM, 10 µM, 100 µM) no inhibition could be observed. The growth rates of yeast cells which were incubated with low concentrations of limonene do not show a difference compared to the negative control (incubation of analogous yeast strains with YPD without limonene).

Fig. 12: Limonene toxicity assay evaluation.

Outlook: Characterization of Enzymatic Activity

Figure 13: Enzymatic assay that will be used to characterize limonene synthase. The principle is well known from pyrosequencing and couples the production of pyrophosphate to the luminescence produced by the firefly luciferase being dependent on ATP.