Metabolic Engineering of the Feverfew Parthenolide Pathway in N. Benthamiana

(1)

Wageningen UR,

Laboratory of Plant

Physiology

Kortooms, Joris

12-06-2014

INTERNSHIP REPORT

M

ETABOLIC

E

NGINEERING

OF

THE

F

EVERFEW

(2)

INTERNSHIP REPORT

M

ETABOLIC

E

NGINEERING OFTHE

F

EVERFEW

P

ARTHENOLIDE

P

ATHWAYIN

N.

BENTHAMIANA

Version 2

Student name:

Joris Kortooms

Supervisor:

Arman Beyraghdar Kashkooli

Secondary supervisors:

Sander van der Krol Lute-Harm Zwiers

Research institute:

Wageningen University and Research Centre Department of Plant Sciences (Radix)

Laboratory of Plant Physiology

Droevendaalsesteeg 1, 6708 PB Wageningen The Netherlands

Educational institute:

Avans University

Academy of Technology in Health and Environment Lovensdijkstraat 61-63, 4818 AJ Breda

The Netherlands

Internship period:

(3)

Preface

I was happy to start an internship at Wageningen UR. One reason was the possibility of continuing my study with a master there. It was also very interesting to do a project in which some techniques were new for me. Moreover, the project fitted well with the minor of my study: biotechnology.

This internship report presents performed research activities I undertook concerning metabolic engineering of the feverfew pathenolide pathway in N. benthamiana during November 2013 to June 2014, at the Laboratory of Plant Physiology (Radix, Wageningen UR), on behalf of Avans University.

I enjoy working in the lab, not only because it is practical work, but also because it is the kind of research that contributes to the broadening of the knowledge of human medical health. This last aspect is of fundamental nature for me. I like the idea of contributing to this field of science.

The office in which I worked was a change for me. The environment was very quiet and everybody was working independently. Fortunately, it did not take a long time to get used to this. Also, some lab activities were new for me, but I quickly learned how to perform those after my supervisor gave a demonstration.

I would like to thank my supervisor Arman for all his help and guidance and also Sander for his advice.

Enjoy reading! Kind regards, Joris Kortooms

(4)

Abstract

Fragments of cytochrome P450 genes (P450s), lipid transfer protein genes (LTPs) and ATP-binding cassette subfamily G transporter genes (ABCGs) were amplified from cDNA of feverfew (Tanacetum parhenium) ovary phase 4 RNA using PCR. P450 gene fragments were cloned into ImpactTim entry vector using the restriction enzymes Not1 and Pac1 and T4 DNA ligase. LTP-12309 and LTP-21667 gene fragments were cloned into TOPO TA entry vector. LTP-14333, LTP-19412, ABCg-3885 and ABCg-7696 gene fragments were cloned into D-TOPO entry vector.

The following positive entry clones (from E. coli) were used for cloning into destination vector via Gateway Technology LR Clonase 2: P450-8272-E1, P450-8595-E4, P450-9025-E2, LTP-14333-E8, LTP-19412-E4, LTP-21667-E2 and ABCg-3885-E5. P450 gene fragments were cloned into pBinPlus vector and LTP- and ABCg-3885 gene fragments were cloned into pB7GW2 vector. The following destination clones were successfully transformed from

E. coli into A. tumefaciens: P450-8272-D1, P450-8595-D3, P450-9025-D4, LTP-14333-D5,

LTP-19412-D2, LTP-21667-D3 and ABCg-3885-D1.

A. tumefaciens cultures containing P450-8595-D3-2 and P450-9025-D4-2 destination

clones, together with costunolide pathway genes, were infiltrated in N. benthamiana leafs. Leaf extracts were analysed using LC-QTOF-MS. Unfortunately, no conclusion can be given about the effect of the agro-infiltration treatments. This is due to the fact that some samples were mixed up. Nevertheless, no derivative compound of costunolide was detected.

1. Introduction

Terpenes are a large and diverse class of secondary metabolites produced by plants. They play a role in plant defence and are used extensively for their aromatic properties. Also, some compounds that have been identified, are being used as pharmaceuticals. One example of this is artemisinin, which is an antimalarial drug, isolated from the plant

(7)

Terpenes are classified according to the size of their carbon skeleton. Sesquiterpenes are one of the terpene classes that are being further investigated for their possible pharmaceutical activity. In this project the sesquiterpene lactone parthenolide, from feverfew (Tanacetum parthenium), plays a central role. The biosynthetic pathway of parthenolide in feverfew has been fully elucidated. However, the pathway could be more extensive than what is proposed at the moment. Metabolic engineering of the pathway with candidate genes in N. benthamiana could possibly lead to the identification of more genes that are involved in the pathway and/or enhance the production of certain sesquiterpene lactones.

Candidate genes are cytochtome P450 genes, lipid transfer protein genes and ATP-binding cassette subfamily G transporter genes. Cytochrome P450 genes are known for their oxidation and thereby modification of organic compounds. More of those genes may participate in the parthenolide pathway. Lipid transfer proteins and ATP-binding cassette subfamily G transporters seem to play a role in the transport of terpenes. Those genes may affect the amount of sesquiterpene lactones that can be accumulated in certain plant organelles.

The research setup of this project was as follows. Candidate genes were chosen from 454 sequencing data of feverfew cDNA. Expression data was available of feverfew genes during several feverfew ovary phases. Candidate genes were selected based on the similarity of their expression profile to that of parthenolide pathway genes. The cloning fragment from each candidate gene was amplified from cDNA using PCR. The cloning fragment was then cloned into an entry vector. After that, the gene fragments from the entry clones were cloned into a destination vector using the LR Clonase recombination reaction. The destination clones were transformed from E. coli into A. tumefaciens. Several Agrobacterium treatments with different gene combinations were injected into N.

benthamiana leafs. Finally, the leafs were analysed with a LC-QTOF-MS to check for the

production of sesquiterpene lactones.

In Chapter 2 the theoretical background is described. The materials and methods are described in Chapter 3. Chapter 4 displays the results. The conclusion and discussion is described in Chapter 5. The recommendations can be found in Chapter 6. The attachments are displayed in the last Chapter.

2. Theoretical Background

2.1. Terpenes

Terpenes are the largest and most widespread class of naturally occurring organic chemicals accounting for more than 40.000 compounds. They are synthesised by plants via secondary metabolism. They have numerous biological activities and are extensively used for their aromatic qualitites. For example, they play a role in plant defence and in the regulation of symbiosis. They do not only play roles in ecology, but they are also used as pharmaceuticals. Some terpenes act as treatment for a human disease, for example,

(8)

artemisin and taxol as malaria and cancer medicines, respectively. Terpenes are also used in ointments due to the fact that they can enhance the uptake of medicinal compounds through the skin owing to the lipophilic character of many terpenes which can change the properties of membranes. [1-5,13,16]

The building blocks of all terpenes are only two five-carbon isomers: isopentenyl diphosphate (IPP, C5) and

dimethylallyl

diphosphate (DMAPP, C5).

One building block is called an isoprene unit. IPP and DMAPP are biosynthesised from mevalonic acid and methylerythritol phosphate which are produced in the cytosol and chloroplasts,

respectively. Addition of IPP to DMAPP or a prenyl diphosphate by

prenyltransferases will generate linear prenyl diphosphates, which serve as precursors for terpene biosynthesis.

Consecutive head-to-tail condensation of IPP will form precursor compounds of different groups of terpenes that contain an increasingly carbon chain length. For instance, the carbon chain length of natural rubber extends to several million atoms. Based on the number of building blocks, terpenes are classified as monoterpenes (C10), sesquiterpenes

(C15) and diterpenes (C20). Geranyl diphosphate, farnesyl diphosphate (FPP) and

geranylgeranyl diphosphate are the precursors of monoterpenes, sesquiterpenes and diterpenes, respectively. Terpenes are synthesized from linear prenyl diphosphates by cyclization reactions (oxidation, reduction, isomerisation, conjugation, etc.) mediated by terpene cyclases, which are known to catalyze the most complex chemical reactions. A simplified scheme of the biosynthesis of terpenes is shown in Figure 1. This project focusses on the sesquiterpene class of terpenes. [1-5]

2.2. Sesquiterpene Lactones from

Asteraceae Species

Among the sesquiterpenes, the sesquiterpene lactones (SLs) are the most prevalent and well-known for their wide variety of biological functions. They have been identified in many plant families and the greatest numbers are found in the Asteraceae subfamily, with over 3000 reported structures. Their structural diversity and diverse potential

(9)

functions as anticancer, anti-inflammatory, anti-malarial, antiviral, antibiotic etc. have made them targets for drug discovery research. Parthenolide, a SL, is responsible for most of the medicinal effects (treatment of fever, migraine and arthritis) that the commonly used herb feverfew (Figure 3) can give as a traditional remedy. [9]_{Another SL is}

artemisinin, which is essential for its antimalarial activity. Also, many plant sesquiterpenes have shown effectiveness against tuberculosis. [3,6,16,17]

SLs are formed by head-to-tail condensation of three isoprene units and subsequent cyclization and oxidative transformation to produce a cis or trans-fused lactone with a 15 carbon backbone. A specific characteristic of SLs is the presence of a γ-lactone ring (closed towards either C-6 or C-8) and an α-methylene group. This project focusses on the SLs found in Asteraceae species. SLs found in this species are classified into the following groups (Figure 2): guaianolides, pseudoguaianolides, germacranolides, and

eudesmanolides. [6,16,17]

The guaianolides and germacranolides from Asteraceae species have been shown to exhibit a wide range of biological

activities. Despite their low molecular weight, the complicated

functionalization of the skeleton of these natural products render them difficult targets for synthesis, especially if large quantities are desired. Biosynthesis is generally far more efficient than the current capabilities of chemical synthesis. For example, ton quantities of 10-deacetyl baccatin III are generated annually in needles of the European yew tree as a fully oxidized renewable starting material for the commercial semi-synthesis of taxol, an anti-cancer drug. Only milligrams of taxol can be chemically synthesized because of lack of general methods, strategies and rules for the functionalization of C–H bonds within complex hydrocarbon systems. Because of their importance for human health a controlled synthesis of the SL skeleton is essential. Moreover, with controlled modifications of this basic skeleton, specific structural variants can be produced, which could help to relate structure to clinical function. [16]

In this project the in planta production of bioactive SLs will be tested by transient expression of feverfew pathway genes. The main target SLs in this project are parthenolide and costunolide (Figure 4), which are both germacranolides. Those two compounds are promising potential drugs for cancer and inflammatory diseases

[6,7]_{. Genetic engineering of SLs and production of novel}

SLs may open up a new point of view in producing novel drugs.

Figure 2: Chemical structure of sesquiterpene lactones in different groups, found in Asteraceae species [16]

Kingdom: Plantae (Plants); Subkingdom: Trachiobionta (Vascular plants); Super division: Spermatophyta (Seed plants); Division: Mangliophyta (Flowering plants); Class: Magnoliopsida (Dicotyledons); Subclass: Asteridae; Order: Asterales; Subfamily: Asteraceae (Aster subfamily); Genus: Tanacetum (tansy); Species: Tanacetum parthenium

Figure 3: Tanacetum parthenium phenotype and botanical classification [9]

Figure 4: Chemical structure of the sesquiterpene lactones parthenolide (left) and costunolide (right) [7]

(10)

2.3. Metabolic Engineering of Sesquiterpene Lactone Biosynthesis

The main precursor of the sesquiterpene biosynthesis is FPP. The initial step of the pathway is the conversion of FPP into germacrene A by germacrene A synthase (GAS). Many GAS genes have been isolated and characterized from a number of plants of the

Asteraceae subfamily (chicory, lettuce, A. annua and feverfew). Germacrene A,

thenceforth, undergoes multiple oxidations to germancrene A alcohol, germacrene A aldehyde and germacrene A acid by germacrene A oxidase (GAO) (found in chicory and inula). Germacrene A acid is hydroxylated by costunolide synthase (COS) and spontaneous cyclisation occurs which thus forms costunolide. Costunolide is then converted to parthenolide by parthenolide synthase (PTS) and/or kauniolide by kauniolide synthase (KAS). Until this point, the proposed pathway has been confirmed (Figure 5). However, there are many compounds that seem to derive from costunolide and kauniolide. It was shown that Tp8886 synthesizes three new compounds (with m/z 248) from costunolide. These compounds have not been identified yet. Moreover, it was shown that Tp8886 uses kauniolide as substrate but again the biosynthesis product (m/z 248) could not yet be identified. In the past twenty years, numerous enzymes and genes involved in terpene biosynthesis have been identified and characterized. Also, in recent years, attempts to introduce terpene cyclase activity in heterologous plants have been successful. At PPH Wageningen University the biosynthesis pathway of the precursor of the sesquiterpene artemisinin (DHAA) was successfully expressed in N. benthamiana. Similarly, using transient expression in N. benthamiana leaves by agro-infiltration of multiple genes encoding the enzymatic steps towards biosynthesis of costunolide, the ectopic production of costunolide was achieved. At present a new gene activity, isolated from feverfew was also characterized which is PTS, the parthenolide synthase. [16]

(11)

Although the biosynthesis pathway of the sesquiterpenes parthenolide and costunolide has been fully elucidated, there are still big steps to be made in metabolic engineering of the pathway in heterologous hosts. Hydroxylation, epoxidation of double bonds and cyclisation of the 10C ring of sequiterpenes are the possible conversions of parthenolide/costunolide, which are mediated by P450 enzymes. The reason for choosing parthenolide and costunolide is that the pathway leading to these two compounds is now elucidated. At present a number of enzymes (GAS, GAO, COS, PTS (Tp.2116) and KAS (Tp.8879)) from the pathway have been characterized (Figure 5). But already this set of enzymes offers opportunities to produce novel SLs, depending on the substrate specificity or promiscuity of the different P450s involved. Also testing pathway-related P450s from feverfew may result in producing novel compounds and branch pathways into finding new SLs, which have not been identified previously. [16]

Plant ATP-binding cassette (ABC) transporters play important roles in the transport of secondary metabolites, such as alkaloids, phenolics, terpenes and wax. The ATP-binding cassette transporter G (ABCg) subfamily may also influence artemisinin accumulation. It was proven that some ABCg transporters are involved in the transport of terpenes or cuticle development in plants and some are induced in the engineered yeast that produced more artemisinin precursors. AaABCG6 and AaABCG7 show a similar expression pattern as two artemisinin biosynthesis-specific genes (amorpha-4,11-diene synthase and the cytochrome P450, CYP71AV1) in different tissues and different leaf developmental stages. If ABCGs from A. annua have any role in transport of artemisinin related compounds, then they may also have an effect on germacrene A related compounds (germacrene A is also present in A. annua trichomes) and thus may have an effect on heterologously produced feverfew sesquiterpenes lactones. [16]

Lipid transfer proteins (LTPs) may also play an important role in the accumulation of pathway products. In many plant species, expression of terpene biosynthesis genes coincides with high expression of LTPs suggesting that these proteins may not only be carriers of lipids but may also be involved in the transport of terpenes. It has been suggested that these proteins may be involved in intracellular trafficking and secretion of terpenes. However molecular evidence for this role has not been provided. LTP genes from feverfew will be tested on the feverfew parthenolide pathway products, as described above for the ABC transporters. [16]

Boosting products of SLs biosynthesis pathway in heterologous hosts is necessary. In PPH Lab of WUR it has been shown that when transient expression of parthenolide biosynthesis without boosting with HMG-CoA reductase (involved in the mevalonate pathway) is done, no free parthenolide can be detected, while boosting with this enzyme resulted in detection of small amounts (2.05 ng·g-1 FW) of free parthenolide in N.

benthamiana leaves. So finding boosting alternatives for parthenolide/costunolide

production for their production and/or further conversion is inevitable. [16]

2.4. Cloning into ImpactTim Entry Vector

In this project ImpactTim entry vector was a vector used for cloning of P450s. ImpactTim is a modified version of ImpactVector1.1 and was created at WUR, PSI. Cloning into the ImpactTim entry vector is done with the use of restriction enzymes. The ImpactTim entry vector contains one Not1 restriction site and one Pac1 restriction site. The enzymes Not1 and Pac1 will cut out a small unimportant part of the vector, due to the fact that the restriction sites lie close to each other. A gene of interest that is cut with the same enzymes, is then able to ligate into the vector between the restriction sites.

(12)

2.5. pCR8/GW/TOPO TA Cloning Kit

Taq polymerase has the ability to add a single deoxyadenosine to the 3’ end of a PCR

product. The cloning kit exploits this by the fact that the linearized TOPO TA entry vector has a single, overhanging deoxythymidine on its 3’ end. This allows the ligation of a PCR fragment into the TOPO TA entry vector. [20]

Topoisomerase 1 from Vaccinia virus binds to double-standed DNA at a “CCCTT” site, which is located on the borders of the gene insertion site, and cleaves the phosphodiester backbone in that strand. This makes room for a gene of interest to be inserted into the vector. The released energy from the broken phosphordiester backbone is used for the formation of a covalent bond between the 3’ phosphate and a thyrosyl residue (Tyr-274) of topoisomerase 1. This bond is subsequently attacked by the 5’ hydroxyl to release the topoisomerase residue. See Figure 6 for an illustration of the reaction. [20]

The TOPO TA entry vector also contains attL1 and attL2 sites which function in the recombination-based transfer of a gene of interest into a destination vector (see Paragraph 2.7). [20] _{In this project the TOPO TA entry vector was}

used for cloning of LTPs.

2.6. pENTR Directional TOPO Cloning Kit

To use this kit, it is first necessary to create a PCR fragment with “CACC” at its 5’ end. This is done by adding “CACC” to the 5’ end of the forward PCR primer. The PCR fragment has to be blunt-ended, so Taq polymerase cannot be used. Instead, a proofreading polymerase is used. A “GTGG” overhang of the D-TOPO entry vector invades the 5’ end of the PCR fragment and replaces “GTGG”. Topoisomerase 1 allows insertion of the PCR fragment into the vector (see Paragraph 2.5). [21]

The TOPO TA entry vector also contains attL1 and attL2 sites which function in the recombination-based transfer of a gene of interest into a destination vector (see Paragraph 2.7). [21] _{In this project the D-TOPO entry vector was used for cloning of LTPs}

and ABCGs.

2.7. Gateway Technology: LR Clonase Enzyme Mix II

The Gateway Technology is a cloning method based on the site-specific recombination system of bacteriophage lambda, which facilitates the integration of lambda into the E.

coli chromosome. Recombination occurs between attachment (att) sites: attB on the E. coli chromosome and attP on the lambda chromosome. Att sites serve as binding sites for

recombination proteins. Upon lambda integration, the attB and attP sites are replaced with attL and attR sites. The crossover occurs between homologous core regions (15 bp) of the att sites. The lambda recombination system is catalyzed by enzymes which are contained in the LR Clonase Enzyme Mix II. The LR reaction facilitates the recombination of an attL substrate (entry clone) with an attR substrate (destination vector) to create an Figure 6: Illustration of TOPO TA cloning

(13)

expression clone with attB sites and a by-product with attP sites. The following proteins are involved in the LR reaction: integrase (lambda), excisionase (lambda) and integration host factors (E. coli). The enzymes work together to bind to the att sites, bring the target sites together, cleave them and covalently attach them to each other. Recombination will then lead to the exchange of att sites between the substrates (entry clone and destination vector) and finally the DNA sequences will be ligated. See Figure 7 for an illustration of the LR reaction. [22] _{In this project LR Clonase II was used for cloning a gene}

from an entry clone into an destination vector to create an expression clone.

2.8. Agro-infiltration in N. benthamiana

Plants have shown to be one of the most promising pharmaceutical production platforms that are robust, scalable, low-cost and safe when compared to mammalian cell culture as being the major platform. Virus-based vectors have shown to allow rapid and high-level transient expression of recombinant proteins in plants. Furthermore, there has been developed a simple, efficient and scalable methodology to introduce a target-gene containing bacterium A. tumefaciens into plant tissue, which is called agro-infiltration. This infiltration process can either be done by the syringe (Figure 8) or the vacuum method. Syringe infiltration is simpler and cheaper, but less robust and less scalable. However, it also allows the flexibility to either infiltrate one target gene, or to introduce multiple target genes on one leaf. Thus, this method can be used for laboratory scale expression of recombinant proteins as well as for comparing different proteins or vectors for yield or expression kinetics. [11]_{In this project N. benthamiana was infiltrated with}

several combinations of target genes. Therefore the syringe method for agro-infiltration was used. N. benthamiana is mostly used as expression host because the plant has a short life cycle, carries relatively large and easily infiltratable leaves that produce recombinant proteins at high levels and the leaf does not show necrosis upon infiltration with most Agrobacterium strains. [10]

Agro-infiltration is based on infiltration of A. tumefaciens cultures into intact plant leaves. The bacterium subsequently transfers a DNA segment, called T-DNA, into the plant cells. In nature this T-DNA is part of the bacterial tumor-inducing (Ti) plasmid that, besides the Figure 7: Illustration of the LR reaction. The “ccdB” gene is used for negative selection. E. coli cells that take up unreacted vectors carrying this gene will die. Recombinated destination vectors will replace the “ccdB” gene with the gene of interest between their attB sites. [22]

(14)

T-DNA, carries the genes required for its transfer. The T-DNA carries effector genes that allow the pathogen to cause crown-gall disease on species in the Rosaceae subfamily. Infiltration of the Ti plasmid in the plant causes an increase in the production of auxins and cytokinins. This makes the cells proliferate rapidly, and results in the formation of a crown gall tumor. In disarmed laboratory strains the effector genes are deleted and the two essential parts of the T-DNA, its left- and right border, are placed on a separate plasmid. Genes placed between these borders will be transferred to the plant. Genes to be transferred may be target genes of interest, selection genes and/or reporter genes. Since the genes required for T-DNA transfer reside of the “helper” plasmid, this system is called the binary vector system. The binary vectors containing the T-DNA can carry inserts of up to over 100 kbp. T-DNA transferred to a plant cell will relocate to the nucleus, where its genes can be transcribed and expressed. The majority of the plant cells in the infiltrated region express the transgene and the expression typically reaches its highest level 2-3 days after infiltration. At later stages the expression is quenched by RNA silencing. [8,15]_{In this project, the agro-infiltration was performed after certain}

candidate genes were cloned into a destination vector and transformed from E. coli into

A. tumefaciens. The used destination vectors for cloning of P450s and for cloning of LTPs

and ABCGs were pBinPlus and pB7GW2, respectively. These vectors are binary T-DNA vectors that can replicate both in E. coli and A. tumefaciens. [23,24]

2.9. Feverfew Candidate Genes

Feverfew candidate genes - P450s, LTPs and ABCGs - were found in the Tanacetum

parthenium database of Terpmed. Terpmed is a project within the “food, agriculture,

fisheries and biotechnology” theme of the European Commision, and is devoted to plant terpenes, with a main focus on sesquiterpene lactones. [14] _{Terpmed was also used as a}

source for 454 sequencing data of the cDNA of feverfew. Beside this there was expression data available of genes related to the candidate genes. First, genes were selected based on the similarity of their expression profile to that of parthenolide pathway genes (see Attachment 1). Secondly, the DNA and protein sequence of the selected genes were aligned with those of known genes which should share high similarity. The final gene candidate filter was checking the amount of amino acids that a selected gene translates and then blasting that sequence to check for desired hits. The candidate genes which were analyzed in this project, including the code names given to them, are shown in Table 1.

Table 1: Feverfew candidate genes analyzed in project

Tanacetum parthenium

gene number

Gene product Gene code

Tp.8272 CYP83B-like P450 enzyme P450-8272 Tp.8595 CYP71A-like P450 enzyme P450-8595 Tp.8630 CYP82A-like P450 enzyme P450-8630 Tp.8743 CYP81B-like P450 enzyme P450-8743 Tp.9025 CYP71D-like P450 enzyme P450-9025 Tp.9177 CYP81B-like P450 enzyme P450-9177

Tp.12309 Lipid transfer protein LTP-12309

Tp.3885 ATP-binding cassette subfamily G transporter protein

ABCg-3885 Tp.7696 ATP-binding cassette subfamily G transporter

protein

(15)

2.10. Project Goal

The goal of this project was to succesfully clone feverfew P450-, LTP- and ABCg genes into

A. tumefaciens and then transiently express them into N. benthamiana. Therefore

different sets of feverfew candidate genes together with parthenolide pathway genes were planned to be infiltrated into N. benthamiana leafs. The next step was the analysis of the production of sesquiterpene lactones in the harvested leafs.

3. Material and Methods

3.1. PCR of Cloning Fragments

PCR cloning fragments were amplified from cDNA, which was created from feverfew ovary phase four RNA with the iScript cDNA Synthesis Kit (Bio-Rad). A 50µl PCR reaction mix was made containing: 10µl of 5X Q5 Reaction Buffer (including 2mM MgCl2, New England

Biolabs), 200µM dNTPs (10mM), 500nM reverse and forward primer (10µM, Integrated DNA Technologies), 50ng cDNA (50ng/µl DNA) or 1µl DNA template, 1U of Q5 High-Fidelity DNA Polymerase (2U/µl, New England Biolabs) and the remaining volume was filled with MQ water. A PCR mix was put in a thermocycler (Applied Biosystems) and a PCR program was executed. See Attachment 2 for the gene codes, used templates, used primers and executed PCR programs. 6X Orange loading buffer was added to the PCR products to 120% of the initial volume. A 0.8% agarose gel (with 0.01% GelRed (Biotium)) was made and PCR products were loaded on the gel, including 1KB Plus DNA Ladder (Life Technologies). The gel was run at 120V for 1 hour. The desired bands were cut out of the gel under UV-light. The DNA of the excised agarose gel bands was extracted with the NucleoSpin Gel & PCR Clean-up Kit (Bioké). The DNA concentration was measured with a Nanodrop instrument.

3.2. Cloning into ImpactTim Entry Vector: P450s

The following 30µl digestion mix was made for ImpactTim vector: 3µl of 10X CutSmart Buffer (New England Biolabs), 2µg ImpactTim entry vector (210.5ng/µl DNA), 5U Pac1 enzyme, 10U Not1-HF enzyme (20U/µl, New England Biolabs) and the remaining volume was filled with MQ water. The following 50µl digestion mix was made for P450-8272 (2), P450-8595 (2), P450-8630 (2), P450-8743 (2), P450-9025 (2) and P450-9177 (2) cloning fragments (see Paragraph 3.1): 5µl of 10X CutSmart Buffer (New England Biolabs), 377.5 to 1270ng PCR fragment (15.1 to 50.8ng/µl DNA), 5U Pac1 enzyme, 10U Not1-HF enzyme (20U/µl, New England Biolabs) and the remaining volume was filled with MQ water. Digestion mixes were incubated at 37°C in a water bath for 1 hour. 6X Orange loading buffer was added to the PCR products to 120% of the initial volume. A 0.8% agarose gel (with 0.01% GelRed (Biotium)) was made and the digestion products were loaded on the gel, including 1KB Plus DNA Ladder (Life Technologies). The gel was run at 120V for 1 hour. The desired bands were cut out of the gel under UV-light. The DNA of the excised agarose gel bands was extracted with the NucleoSpin Gel & PCR Clean-up Kit (Bioké). The DNA concentration was measured with a Nanodrop instrument.

(16)

The following 20µl ligation mix was made for the DNA products after the restriction digest (see text above): 2µl of 5X T4 DNA Ligase Buffer (Life Technologies), 50µg digested ImpactTim entry vector (34.5 ng/µl DNA), 50µg digested P450 cloning fragment (4.6 to 21.4 ng/µl DNA), 1U of T4 DNA Ligase and the remaining volume was filled with MQ water. The molar ratio of the P450 fragments with the ImpactTim entry vector was 3 to 1. Ligation mixes were incubated at 37°C in a water bath for 1 hour. 5µl Ligation mix was added to 50µl chemical competent DH5a E. coli cells. Sample was mixed by stirring with the tip. Samples were incubated on ice for 24 minutes. Cells were heat-shocked at 42°C in a water bath for 30 seconds. 1ml Of room temperature LB medium was added to the samples. Samples were incubated at 37°C with 300RPM shaking for 73 minutes. 50µl Sample was spread out on a LB agar plate containing 100µg/ml kanamycin. Samples were spun down at 4000RPM in a microcentrifuge for 3 minutes. 850µl Medium was removed from the samples. The remaining medium was mixed with the formed pellet by pipetting up and down. 50µl (more concentrated) Sample was spread out on another LB agar plate containing 100µg/ml kanamycin. Plates were incubated at 37°C for 19 hours.

3.3. Cloning into TOPO TA Entry Vector: LTP-12309 & LTP-21667

The following 10µl adding adenosine mix was made for LTP-12309 cloning fragment (see Paragraph 3.1) 236ng LTP-12309 fragment (47.2ng/µl DNA), 1mM dATP (10mM), 1µl of 10x SuperTaq Buffer (HT Biotechnology), 1.5mM MgCl2 (50mM), 0.5U SuperTaq enzyme

(5U/µl, HT Biotechnology) and the remaining volume was filled with nuclease-free water. Samples were incubated at 72°C for 30 minutes. The DNA concentration was measured with a Nanodrop instrument. The following 50µl adding adenosine mix was made for 21667 cloning fragment (see gene codes in Attachment 2 for the explanation): 417ng LTP-21667 fragment (83.4ng/µl DNA), 200µM dATP (10mM), 5µl of 10x SuperTaq Buffer (HT Biotechnology), 1.5mM MgCl2 (50mM), 1U SuperTaq enzyme (5U/µl, HT Biotechnology)

and the remaining volume was filled with nuclease-free water. Samples were incubated at 72°C for 20 minutes. The 3’A LTP-21667 product was purified with the NucleoSpin Gel & PCR Clean-up Kit (Bioké). The DNA concentration was measured with a Nanodrop instrument.

The following 6µl TOPO TA cloning reaction was made for LTP-12309 and LTP-21667 cloning fragments with 3’A (see text above): 94.4 ng LTP-12309 fragment (from adding adenosine mix, 23.6ng/µl) or 85.6ng LTP-21667 fragment (from purified adding adenosine mix, 21.4ng/µl), 1µl diluted salt solution (300mM NaCl, 15mM MgCl2) and 1µl TOPO TA

entry vector [20]_{. The molar ratio of the LTP-12309 fragment and LTP-21667 fragment with}

the TOPO TA entry vector was 26 to 1 and 41 to 1, respectively. Both reactions were incubated at room temperature for 5 minutes. Samples were placed on ice and 3µl of TOPO TA reaction was added to 50µl electrocompetent DH5a E. coli cells. Sample was mixed by stirring with the tip. Samples were incubated on ice for 30 minutes. Cells were electroporated at 1.8V, 25µF and 200 Ohm in a 1mm cuvette. 250µl S.O.C. medium (Life Technologies) was added to the sample in the cuvette. Samples were put in the original tubes and were incubated at 37°C for 1 hour with gently shaking by hand every 20 minutes. 10µl, 20µl And 50µl sample was spread out on LB agar plates containing 100µg/ml spectinomycin. Plates with the LTP-12309 sample were incubated at 37°C for 3 days and plates with the LTP-21667 sample were incubated at 37°C for 1 day.

(17)

3.4. Cloning into D-TOPO Entry Vector: LTP-14333, LTP-19412, ABCg-3885 &

ABCg-7696

The following 6µl D-TOPO cloning reaction was made on ice for LTP-14333 (2), LTP-19412 (2), ABCg-3885 (4) and ABCg-7696 cloning fragments (see Paragraph 3.1): 9.9ng LTP-14333 fragment (19.8ng/µl) or 8.8ng LTP-19412 fragment (17.6ng/µl) or 33.1ng ABCg-3885 fragment (18.9ng/µl) or 33.8ng ABCg-7696 fragment (33.8 ng/µl), 1µl salt solution (1.2M NaCl, 60mM MgCl2), 3.5µl MQ water and 1µl D-TOPO entry vector [21]. The molar

ratio of the LTP-14333-, LTP-19412-, ABCg-3885 and ABCg-7696 fragments with the D-TOPO entry vector was 2 to 1, 3 to 1, 2 to 1 and 1 to 1, respectively. Samples were incubated at 23°C for 30 minutes. Samples were places on ice and 2µl D-TOPO reaction was added to 50µl chemical competent DH5a E. coli cells. Sample was mixed by gently flicking the tube. Samples were incubated on ice for 5 minutes. Cells were heat-shocked at 42°C in a water bath for 30 seconds. The samples were put on ice and 250µl of room temperature S.O.C. medium (Life Technologies) was added to the samples. Samples were incubated at 37°C with 300RPM shaking for 1 hour. 50µl And 200µl sample was spread out on LB agar plates containing 50µg/ml kanamycin. Plates were incubated at 37°C for 19 hours.

3.5. Restriction Digests: P450s & LTP-21667

The following 30µl digestion mix was made to check the quality of the first few P450 entry clones in ImpactTim vector (see Paragraph 3.2 and 3.9): 3µl of 10X CutSmart Buffer (New England Biolabs), 2µg P450 plasmid (263.6 to 344.4ng/µl DNA), 5U Pac1 enzyme, 10U

Not1-HF enzyme (20U/µl, New England Biolabs) and the remaining volume was filled with

MQ water. The following 25µl digestion mix was made for all ImpactTim P450 entry clones (see Paragraph 3.2): 2.5µl of 10X CutSmart Buffer (New England Biolabs), 1054.4 to 1562ng P450 plasmid (263.6 to 390.5ng/µl DNA), 5U Pac1 enzyme, 10U Not1-HF enzyme (20U/µl, New England Biolabs) and the remaining volume was filled with MQ water. Digestion mixes were incubated at 37°C in a water bath for 1 hour. 6X Orange loading buffer was added to the PCR products to 120% of the initial volume. A 0.8% agarose gel (with 0.01% GelRed (Biotium)) was made and the digestion products were loaded on the gel, including 1KB Plus DNA Ladder (Life Technologies). The gel was run at 120V for 1 hour.

The following 25µl digestion mix was made to check the quality of two TOPO TA-LTP-21667-entry clones (see Paragraph 3.3 and 3.9): 3µl of 10X CutSmart Buffer (New England Biolabs), 0.5µg LTP plasmid (119.9 to 174.1ng/µl DNA), 0.5µl EcoRI-HF enzyme or

EcoRV-HF enzyme (20U/µl, New England Biolabs) and the remaining volume was filled

with MQ water. Digestion mixes were incubated at 37°C in a water bath for 75 minutes. 6X Orange loading buffer was added to the PCR products to 120% of the initial volume. A 0.8% agarose gel (with 0.01% GelRed (Biotium)) was made and the digestion products were loaded on the gel, including 1KB Plus DNA Ladder (Life Technologies). The gel was run at 120V for 1 hour.

The following 50µl digestion mix was made for linearization of entry clone LTP-21667-E2 (see Paragraph 3.3 and 3.9): 5µl of 10X CutSmart Buffer (New England Biolabs), 1741ng LTP-21667-E2 (174.1ng/µl DNA), 1µl EcoRV-HF enzyme (20U/µl, New England Biolabs) and the remaining volume was filled with MQ water. Digestion mix was incubated at 37°C in a water bath for 65 minutes. 6X Orange loading buffer was added to the PCR product to 120% of the initial volume. A 0.8% agarose gel (with 0.01% GelRed (Biotium)) was made

(18)

and the digestion product was loaded on the gel, including 1KB Plus DNA Ladder (Life Technologies). The gel was run at 120V for 1 hour. The band looked good on the gel so the DNA of the digestion product was purified with the NucleoSpin Gel & PCR Clean-up Kit (Bioké). The DNA concentration was measured with a Nanodrop instrument.

3.6. Analysis of ABCg-7696 Entry Clones

The following 24µl mix was made for putting ABCg-7696 entry clones (see Paragraph 3.4 and 3.9) on gel to check their quality: 201.1 to 209.6ng ABCg-7696 entry clone (46.1 to 154,7ng/µl DNA), 4µl of 6X orange loading buffer and the remaining volume was filled with MQ water. A 0.8% agarose gel (with 0.01% GelRed (Biotium)) was made and samples were loaded on the gel, including 1KB Plus DNA Ladder (Life Technologies). The gel was run at 120V for 1 hour.

The following 50µl PCR mix was made to check the quality of ABCg-7696 entry clones (see Paragraph 3.4 and 3.9): 10µl of 5X Q5 Reaction Buffer (including 2mM MgCl2, New

England Biolabs), 200µM dNTPs (10mM), 500nM reverse and forward primer (10µM, Integrated DNA Technologies), 1µl ABCg-7696 entry clone (46.1 to 154,7ng/µl DNA), 1U of Q5 High-Fidelity DNA Polymerase (2U/µl, New England Biolabs) and the remaining volume was filled with MQ water. A PCR mix was put in a thermocycler (Applied Biosystems) and the following PCR program was started: 30 seconds at 98°C, 10 seconds at 98°C (30 cycles), 30 seconds at 62°C (30 cycles), 2 minutes and 30 seconds at 72°C (30 cycles) and 2 minutes at 72°C. 6X Orange loading buffer was added to the PCR products to 120% of the initial volume. A 0.8% agarose gel (with 0.01% GelRed (Biotium)) was made and PCR products were loaded on the gel, including 1KB Plus DNA Ladder (Life Technologies). The gel was run at 120V for 1 hour.

The following 20µl digestion mix was made to check the quality of ABCg-7696 entry clones (see Paragraph 3.4 and 3.9): 2µl of 10X CutSmart Buffer (New England Biolabs), 0.5µg ABCg-7696 plasmid (46.1 to 154,7ng/µl DNA), 5U Not1-HF enzyme (20U/µl, New England Biolabs), 5U Asc1 enzyme (10U/µl, New England Biolabs) and the remaining volume was filled with MQ water. Digestion mixes were incubated at 37°C in a water bath for 1 hour. 6X Orange loading buffer was added to the PCR products to 120% of the initial volume. A 0.8% agarose gel (with 0.01% GelRed (Biotium)) was made and the digestion products were loaded on the gel, including 1KB Plus DNA Ladder (Life Technologies). The gel was run at 120V for 1 hour.

3.7. Cloning into Destination Vector with LR Clonase 2: 8272,

P450-8595, P450-9025, LTP-14333, LTP-19412, LTP-21667 & ABCg-3885

The following 8µl mix was made at room temperature for entry clones 8272, P450-8595, P450-9025 (see Paragraph 3.2, 3.3, 3.4 and 3.9): 344.3 ng P450-8272-E1 (344.4ng/µl DNA) or 325ng P450-8595-E4 (325ng/µl DNA) or 343.9ng P450-9025-E2 (343.9ng/µl DNA), 237.2ng pBinPlus destination vector (see Attachment 5, 237.2ng/µl DNA) and the remaining volume was filled with TE buffer (pH 8.0). The following 8µl mix was made at room temperature for entry clone LTP-21667 (3.9. Plasmid Extraction & Colony PCR): 150.3ng of linearized LTP-21667-E2 (see Paragraph 3.5) (41.3ng/µl DNA), 150.2ng pB7GW2 destination vector (see Attachment 6, 91.6ng/µl DNA) and the

(19)

remaining volume was filled with TE buffer (pH 8.0). The following 8µl mix was made at room temperature for entry clones LTP-14333, LTP-19412 and ABCg-3885 (3.9. Plasmid Extraction & Colony PCR): 158.3ng LTP-14333-E8 (158.3ng/µl DNA) or 142.8ng LTP-19412-E4 (142.8ng/µl DNA) or 224.2ng ABCg-3885-E5 (224.2ng/µl DNA), 163.5ng pB7GW2 destination vector (see Attachment 6, 163.5ng/µl DNA) and the remaining volume was filled with TE buffer (pH 8.0). LR Clonase Enzyme Mix 2 [22]_{was mixed by vortexing briefly}

twice. 2µl Of the enzyme mix was added to each sample. LR reactions were incubated at 25°C for 1 hour. 1µl Proteinase K was added to each LR reaction. Samples were incubated at 37°C for 10 minutes. The whole LR reaction was added to 50µl chemical competent DH5a E. coli cells. Samples were mixed by gently flicking the tube. Samples were incubated on ice for 30 minutes. Cells were heat-shocked at 42°C in a water bath for 30 seconds. The samples were put on ice and 250µl of room temperature S.O.C. medium (Life Technologies) was added to the samples. Samples were incubated at 37°C with 300RPM shaking for 1 hour. 50µl And 200µl sample was spread out on LB agar plates containing the appropriate antibiotic. P450s: 100µg/ml kanamycin; LTPs and ABCg-3885: 100µg/ml spectinomycin. Plates were incubated at 37°C for 19 hours.

3.8. Transformation into A. tumefaciens: 8272, 8595,

P450-9025, LTP-14333, LTP-19412, LTP-21667 & ABCg-3885

The following amount of each destination clone (see Paragraph 3.7 and 3.9) was put into 50µl electrocompetent AGL0 A.tumefaciens cells: 37ng P450-8272-D1 (73.9ng/µl DNA) or 90.4ng P450-8595-D3 (90.4ng/µl DNA) or 89.6ng P450-9025-D4 (89.6ng/µl DNA) or 82.5ng LTP-14333-D5 (164.9ng/µl DNA) or 106.8ng LTP-21667-D3 (106.8ng/µl DNA). Samples were mixed by gently flicking the tube. Samples were incubated on ice for 7 minutes. Cells were electroporated at 1.8V, 25µF and 200 Ohm in a 1mm cuvette. 250µl S.O.C. medium (Life Technologies) was added to the sample in the cuvette. Samples were put in the original tubes and were incubated at 28°C at 300RPM for 1 hour. 20µl And 100µl sample was spread out on LB agar plates containing the appropriate antibiotic. All genes: 25µg/ml rifampicin; P450s: 100µg/ml kanamycin; LTPs and ABCg-3885: 100µg/ml spectinomycin. Exception for P450-8272-D1 and LTP-14333-D5: 10µl sample plus 40µl S.O.C. medium and 20µl sample plus 30µl S.O.C. medium was spread out on plates. Plates were incubated at 28°C for 37 hours. A single colony was taken from a plate and the tip was put in ~5ml LB medium containing the appropriate antibiotic. This culture was incubated at 28°C with 300 RPM shaking for 46 hours. After incubation, 300µl culture was mixed with 300µl of 50% glycerol to create a glycerol stock which was subsequently put in a -80°C freezer. The plasmid was extracted from the remaining culture with the QIAprep Spin Miniprep Kit (Qiagen). The DNA concentration was measured with a Nanodrop instrument.

3.9. Colony PCR & Plasmid Extraction

A single colony (a clone) was taken from a plate (see Paragraph 3.2, 3.3, 3.4 and 3.7) and was mixed with 12µl MQ water. 2µl Of this mixture was pipetted in ~5ml LB medium containing the appropriate antibiotic (depending on the vector used for cloning). This culture was incubated at 37°C with 300 RPM shaking for ~1 day. After incubation, the plasmid was extracted from the culture with the QIAprep Spin Miniprep Kit (Qiagen). The DNA concentration was measured with a Nanodrop instrument.

(20)

A 50µl PCR reaction mix was made containing: 12µl colony mixture (see text above), 5µl of 10x SuperTaq Buffer (HT Biotechnology), 200µM dNTPs (10mM), 400nM reverse and forward primer (10µM, Integrated DNA Technologies), 1.25U of SuperTaq enzyme (5U/µl, HT Biotechnology) and the remaining volume was filled with MQ water. A PCR mix was put in a thermocycler (Applied Biosystems) and a PCR program was executed. See Attachment 3 for the used templates, primers and PCR programs. 6X Orange loading buffer was added to the PCR products to 120% of the initial volume. A 0.8% agarose gel (with 0.01% GelRed (Biotium)) was made and PCR products were loaded on the gel, including 1KB Plus DNA Ladder (Life Technologies). The gel was run at 120V for 1 hour.

3.10. Sequencing of Clones

5µl Of a clone (see Paragraph 3.2, 3.3, 3.4, 3.7 and 3.9) and 5µl of an appropriate primer (5µM) were pipetted in a sample tube. The concentration of some clones needed to be increased. This was done by evaporating a bit of those clones with nitrogen flow. Samples were send to Macrogen for sequencing of the samples.

3.11. PCR of Cloning Fragments from A. tumefaciens Clones: P450-8272,

P450-8595, P450-9025, LTP-14333, LTP-19412, LTP-21667 & ABCg-3885

A 50µl PCR reaction mix was made containing: 10µl of 5X Q5 Reaction Buffer (including 2mM MgCl2, New England Biolabs), 200µM dNTPs (10mM), 500nM reverse and forward

primer (10µM, Integrated DNA Technologies), 1µl A. tumefaciens destination clone (see Paragraph 3.8), 1U of Q5 High-Fidelity DNA Polymerase (2U/µl, New England Biolabs) and the remaining volume was filled with MQ water. A PCR mix was put in a thermocycler (Applied Biosystems) and a PCR program was executed. See Attachment 4 for the used templates, primers and PCR programs. 6X Orange loading buffer was added to the PCR products to 120% of the initial volume. A 0.8% agarose gel (with 0.01% GelRed (Biotium)) was made and PCR products were loaded on the gel, including 1KB Plus DNA Ladder (Life Technologies). The gel was run at 120V for 1 hour.

3.12. Agro-infiltration in N. benthamiana: P450-8595 & P450-9025

A. tumefaciens cultures (see Paragraph 3.8) containing P450-8595-D3-2, P450-9025-D4-2, AtHMGR, TpGAS 1.5, CiGAO, CiCOS and P19 genes and empty vector (pBinPlus

destination vector (see Attachment 5)) were made by taking a bit of frozen glycerol stock with a pipette tip and putting the tip in ~5ml LB medium containing 25µg/ml rifampicin and 100µg/ml kanamycin. Cultures were incubated at 28°C with 300 RPM shaking for 23 hours. LB medium again containing 25µg/ml rifampicin and 100µg/ml kanamycin was added to all cultures to ~12.5ml. Again, cultures were incubated at 28°C with 300 RPM shaking for 18 hours. Remark: for CiCOS and P19 a 4 days old culture (including 50 hours incubation at 28°C with 300 RPM shaking) was used, instead of 2 days. Cultures were spun down at 3500RPM in a table top centrifuge for 15 minutes. The medium was removed from all pellets. 1L Agro-infiltration buffer was made: 10mM MgCl2 (500mM),

10mM MES-KOH (500mM), 0.1mM acetosyringone (100mM) and the remaining volume was filled with demi water. Pellets were resuspended in ~20 ml agro-infiltration buffer. The OD600 of the samples was measured with a spectrophotometer. Agro-infiltration buffer was added to a volume of at least 50 ml till the samples had an OD600 of ~0.5.

(21)

Different gene treatment mixes were made: 1: 5ml AtHMGR, 5ml TpGAS, 5ml P19, 5ml P450-8595-D3-2 and 5ml P450-9025-D4-2; 2: 5ml AtHMGR, 5ml TpGAS, 5ml P19 and 10ml empty vector; 3: 5ml AtHMGR, 5ml TpGAS, 5ml CiGAO, 5ml CiCOS, 5ml P19, 5ml P450-8595-D3-2 and 5ml P450-9025-D4-2; 4: 5ml AtHMGR, 5ml TpGAS, 5ml CiGAO, 5ml P19 and 10ml empty vector; 5: AtHMGR, 5ml TpGAS, 5ml CiGAO, 5ml P19, 5ml P450-8595-D3-2 and 5ml P450-90P450-8595-D3-25-D4-P450-8595-D3-2; 6: 5ml AtHMGR, 5ml TpGAS, 5ml CiGAO, 5ml CiCOS, 5ml P19 and 10ml empty vector. Gene treatment mixes were mixed on a roller bench for 2 hours. One leaf per plant and 3 plants in total (3 replicates) were infiltrated per treatment. The infiltrated leafs (18x) were harvested 5 days after agro-infiltration.

3.13. Grinding of Leafs & Extracting Metabolites

Leafs (see Paragraph 3.12) were grinded to a fine powder in liquid nitrogen with a pestle and mortar. Sample tubes were filled with ~100mg leaf powder. 300µl Methanol was added to each sample. Samples were mixed by vortexing. Leaf material was sonicated for 15 minutes. Samples were spun down at 17968xg in a table top centrifuge for 15 minutes. Replicates (3x) of each agro-infiltration treatment were put together by pipetting 200µl of each replicate in one tube. Samples (6x) were mixed by vortexing. 200µl Of each sample was filtered into a LC-QTOF-MS sample tube.

3.14. Analysing Leaf Extracts with LC-QTOF-MS

For LC-QTOF-MS analysis of leaf extracts (see Paragraph 3.13), a Waters Alliance 2795 HPLC connected to a Waters 2996 PDA detector and subsequently a QTOF Ultima V4.00.00 mass spectrometer (Waters, MS technologies, UK) operating in positive ionization mode was used. An analytical column (Luna 3 µ C18/2 100A; 2.0 x 150mm; Phenomenex, USA) attached to a C18 pre-column (2.0 x 4mm; Phenomenex, USA) was used. Degassed eluent A [ultra-pure water: formic acid (1000:1, v/v)] and eluent B [acetonitril:formic acid (1000:1, v/v)] were used at a flow rate of 0.19 ml min-1_{. Masses}

were recorded between m/z 80 and m/z 1500; leucine enkaphalin ([M-H]-_{=554.2620) was}

used as a lock mass for on-line accurate mass correction. The gradient of the HPLC started at 5% eluent B and increased linearly to 75% eluent B in 45 minutes, after which the column was washed and equilibrated for 15 minutes before next injection. Injection volume was 5µl.

(22)

4. Results

4.1. P450s

4.1.1. PCR of P450 Cloning Fragments

Figure 9: Gel of P450 cloning fragments. Lane 1: 1KB Plus DNA Ladder; Lane 2: P450-8595 (2) (expected size: 1565bp); Lane 3: P450-9025 (2) (expected size: 1502bp); Lane 4: P450-8743 (2) (expected size: 1502bp); Lane 5: P450-8272 (2) (expected size: 1565bp); Lane 6: P450-8630 (2) (expected size: 1409bp); Lane 7: P450-9177 (2) (expected size: 1529bp); Lane 8: negative control (no template and no primers). Arrows: P450 cloning fragments.

Figure 9 displays P450 cloning fragments on gel. All the P450 cloning fragments show a size that is close to the expected size. The arrows in Figure 9 indicate the bands that were cut out of the gel. (See Paragraph 3.1.)

4.1.2. Restriction Digest of ImpactTim Entry Vector and P450 Cloning Fragments

Figure 10: Gel of digested P450 cloning fragments. Lane 1: 1KB Plus DNA Ladder; Lane 2: ImpactTim entry vector (expected size: 5363bp); Lane 3: P450-9025 (expected size: 1502bp); Lane 4: P450-8630 (expected size: 1409bp); Lane 5: P450-9177 (expected size: 1529bp); Lane 6: P450-8272 (expected size: 1565bp); Lane 7: P450-8743 (expected size:

(23)

1502bp); Lane 8: P450-8595 (expected size: 1565bp); Lane 9: negative control (no template). Arrows: P450 cloning fragments.

Figure 10 displays digested ImpactTim entry vector and P450 cloning fragments on gel. All the fragments show a size that is close to the expected size. The bands of lane 2 to 8 in Figure 10 were cut out of the gel. (See Paragraph 3.2.)

4.1.3. Colony PCR of P450 entry clones in ImpactTim Vector

Table 2: Results of cloning P450 fragments into ImpactTim vector and colony PCR results

Gene code Amount of plates Total amount of colonies Amount of colonies tested with colony PCR Amount of positive colonies P450-8272 2 26 2 2 P450-8595 1 5 5 5 P450-8630 2 24 7 7 P450-8743 1 3 3 3 P450-9025 2 15 2 2 P450-9177 2 8 2 2

Table 2 displays the results after cloning the P450 fragments into ImpactTim vector and after colony PCR. All the tested entry clones were positive. (See Paragraph 3.2 and 3.9)

4.1.4. Restriction Digests of P450 Entry Clones

Figure 11: Gel of digested P450 entry clones in ImpactTim vector. Lane 1: 1KB Plus DNA Ladder; Lane 2: P450-8630-E1 (expected sizes: 5363bp (ImpactTim vector) and 1409bp (P450-8630)); Lane 3: P450-9177-E1 (expected sizes: 5363bp (ImpactTim vector) and 1529bp (P450-9177)); Lane 4: P450-8272-E1 (expected sizes: 5363bp (ImpactTim vector) and 1565bp (P450-8272)); Lane 5: P450-8595-E1 (expected sizes: 5363bp (ImpactTim vector) and 1565bp (P450-8595)); Lane 6: P450-9025-E1 (expected sizes: 5363bp (ImpactTim vector) and 1502bp (P450-9025)); Lane 7: P450-8743-E1 (expected sizes: 5363bp (ImpactTim vector) and 1502bp (P450-8743)); Lane 8: negative control (no template). Arrows: P450 cloning fragments.

(24)

Figure 11 displays digested P450 entry clones in ImpactTim vector on gel. The visible bands show a size that is close to the expected size. However, the P450-8630, P450-8595 and P450-8743 fragments are not visible. (See Paragraph 3.5.)

Figure 12: Gel of digested P450 entry clones in ImpactTim vector. Lane 1,15,16,30: 1KB Plus DNA Ladder; Lane 4,5: P450-8272-E2,E1 (expected sizes: 5363bp (ImpactTim vector) and 1565bp (P450-8272)); Lane 6-10: P450-8595-E1,E2,E3,E4,E5 (expected sizes: 5363bp (ImpactTim vector) and 1565bp (P450-8595)); Lane 11-14,19-22:

P450-8630-E2,E3,E4,E5,E1,E6,E7,E8 (expected sizes: 5363bp (ImpactTim vector) and 1409bp (P450-8630)); Lane 23-25: P450-8743-E2,E3,E1 (expected sizes: 5363bp (ImpactTim vector) and 1502bp (P450-8743)); Lane 26,27: P450-9025-E2,E1 (P450-9025-E1 (expected sizes: 5363bp (ImpactTim vector) and 1502bp (P450-9025)); Lane 28,29: P450-9177-(P450-9025-E1,E2 (expected sizes: 5363bp (ImpactTim vector) and 1529bp (P450-9177)). Arrows: P450 cloning fragments.

Figure 12 displays digested P450 entry clones in ImpactTim vector on gel. The visible bands show a size that is close to the expected size. However, the P450-8272 (2 clones) and P450-8630 (8 clones) fragments are not visible. (See Paragraph 3.5.)

4.1.5. PCR of P450 Destination Clones in pBinPlus Vector

Table 3: Results of cloning P450 fragments into pBinPlus vector and colony PCR results.

Gene code Amount of plates

Total amount of colonies

Amount of colonies tested with PCR Amount of positive colonies P450-8272 E. coli 2 >100 4 (colony PCR) 4 P450-8595 E. coli 2 >100 4 (colony PCR) 4 P450-9025 E. coli 2 >150 4 (colony PCR) 4

(25)

P450-8272 A. tumefaciens 2 1 1 (normal PCR) 1

P450-8595 A. tumefaciens 2 >1000 4 (normal PCR) 4

P450-9025 A. tumefaciens 2 >1000 4 (normal PCR) 4

Table 3 displays the results after cloning the P450 fragments into pBinPlus vector and after colony PCR. All the tested destination clones were positive. (See Paragraph 3.7, 3.8, 3.9 and 3.11.)

4.1.6. Sequencing Results of P450 Clones

Table 4: Sequencing results of P450 entry- and destination clones.

Gene code (E: entry clone; D: destination clone) Amount of the gene sequenced

P450-8272-E1 with primer F4 98%

P450-9025-E1 with primer FD 0%

P450-8272-D1 with primer F3 98%

P450-8272-D1 (conc increased) with primer F3 89%

Table 4 displays the sequencing results of P450 entry- and destination clones. Most clones share high similarity with their respective gene, except for P450-8743, P450-9025 and P450-9177, which show no similarity at all. (See Paragraph 3.10.)

4.1.7. Results of Agro-infiltration of P450-8595-D3-2 and P450-9025-D4-2 Genes together with Parthenolide Pathway Genes in N. benthamiana

25

Figure 13: LC-QTOF-MS chromatographs of N. benthamiana leaf compounds. Leafs were infiltrated with two different gene mix treatments. Chromatograph above: AtHMGR + TpGAS + CiGAO + CiCOS + P19 + P450-8595-D3-2 + P450-9025-D4-2; Chromatograph below: AtHMGR + TpGAS + CiGAO + P19 + P450-8595-D3-2 + P450-9025-D4-2. Circles: costunolide

(26)

Figure 13 displays two chromatographs of N. benthamiana leaf compounds. The results, however, are not as expected. Costunolide conjugates should have been detected in the above chromatograph of Figure 13, because the gene mix treatment contained

costunolide synthase (CiCOS). And on the other hand, costunolide conjugates should not have been detected in the below chromatograph, because the gene mix treatment did not contain CiCOS. Also, no derivative compound of costunolide was detected (See Paragraph 3.12, 3.13 and 3.14.)

4.2. LTPs

4.2.1. PCR of LTP Cloning Fragments

Figure 14, 15 and 16 display LTP cloning fragments on gel. All cloning fragments show a size that is close to the expected size. The arrows in Figure 14, 15 and 16 indicate the bands that were cut out of the gel. (See Paragraph 3.1.)

Figure 15: Gel of LTP-14333 and LTP-19412 cloning fragments. Lane 1: 1KB Plus DNA Ladder; Lane 3: LTP-14333 (expected size: 602bp); Lane 5: LTP-19412 (expected size: 365bp). Blue arrow: LTP-14333 cloning fragment; Green arrow: LTP-19412 cloning fragment.

Figure 14: Gel of LTP-12309 cloning fragment. Lane 1: 1KB Plus DNA Ladder; Lane 2,3: LTP-12309 (expected size: 1059bp). Arrows: LTP-12309 cloning fragment.

Figure 16: Gel of LTP-21667 cloning fragment. Lane 1: 1KB Plus DNA Ladder; Lane 2,3: LTP-21667 (expected size: 580bp). Arrows: LTP-21667 cloning fragment.

(27)

4.2.2. Colony PCR of LTP Entry Clones in TOPO TA or D-TOPO Vector

Table 5: Results of cloning LTP fragments into TOPO TA or D-TOPO vector and colony PCR results.

Gene code Amount of plates

Total amount of colonies Amount of colonies tested with colony PCR Amount of positive colonies LTP-12309 (TOPO TA) 3 4 3 3 LTP-14333 (D-TOPO) 2 >100 5 4 LTP-19412 (D-TOPO) 2 >100 5 5 LTP-21667 (TOPO TA) 3 4 2 2

Table 5 displays the results after cloning the LTP fragments into TOPO TA or D-TOPO vector and after colony PCR. All the tested entry clones were positive. (See Paragraph 3.3, 3.4 and 3.9.)

4.2.3. Restriction Digest of LTP-21667 Entry Clones

Figure 17: Gel of digested LTP-21667 entry clones in TOPO TA vector. Lane 1: 1KB Plus DNA Ladder; Lane 3: LTP-21667-E1 cut with EcoR5-HF (expected size: 2397bp (TOPO TA vector plus LTP-21667)); Lane 4: LTP-21667-E1 cut with EcoR1-HF (expected sizes: 2817bp (TOPO TA vector) and 580bp (LTP-21667)); Lane 5: LTP-21667-E2 cut with EcoR5-HF (expected size: 2397bp (TOPO TA vector plus LTP-21667)); Lane 6: LTP-21667-E2 cut with EcoR1-HF (expected sizes: 2817bp (TOPO TA vector) and 580bp (LTP-21667)). Arrows: LTP-21667 cloning fragment.

Figure 17 displays digested LTP-21667 entry clones in TOPO TA vector on gel. The visible bands show a size that is close to the expected size. (See Paragraph 3.5.)

4.2.4. PCR of LTP Destination Clones in pB7GW2 Vector

Table 6: Results of cloning LTP fragments into pB7GW2 vector and colony PCR results.

Gene code Amount of plates Total amount of colonies Amount of colonies tested with PCR Amount of positive colonies LTP-14333 E. coli 2 >250 5 (colony PCR) 4 LTP-19412 E. coli 2 >100 5 (colony PCR) 5 LTP-21667 E. coli 2 >150 4 (colony PCR) 4 LTP-14333 A. tumefaciens 2 >1000 2 (normal PCR) 2 LTP-19412 A. tumefaciens 2 >1000 2 (normal PCR) 2

(28)

LTP-21667 A. tumefaciens 2 >1000 2 (normal PCR) 2

Table 6 displays the results after cloning the LTP fragments into pB7GW2 vector and after colony PCR. Most of the tested destination clones were positive. (See Paragraph 3.7, 3.8, 3.9 and 3.11.)

4.2.5. Sequencing Results of LTP Clones

Table 7: Sequencing results of LTP entry- and destination clones.

LTP-14333-E5 with primer R4 91%

LTP-14333-E8 with primer F4 97%

LTP-19412-E5 with primer R4 90%

LTP-14333-D4 with primer F4 95%

LTP-14333-D5 (conc increased) with primer F4 95%

LTP-21667-D3 with primer F1 99.7%

LTP-21667-D3 (conc increased) with primer F1 97%

Table 7 displays the sequencing results of LTP entry- and destination clones. Every clone shares high similarity with their respective gene. (See Paragraph 3.10.)

4.3. ABCGs

4.3.1. PCR of ABCg Cloning Fragments

Figure 18 and 19 display ABCg cloning fragments on gel. On lane 7 In Figure 17 the band shows a size close to the expected size of the ABCg-3885 cloning fragment. On lane 2 and 3 in Figure 19 the bands should represent the ABCg-7696 cloning fragment, although the bands are ~500bp smaller than expected. The arrows in Figure 18 and 19 indicate the bands that were cut out of the gel. (See Paragraph 3.1.)

Figure 19: Gel of ABCg-7696 cloning fragment. Lane 1: 1KB Plus DNA Ladder; Lane 2,3: ABCg-7696 (expected size: 4302bp). Arrows: ABCg-7696 cloning fragment. Figure 18: Gel of ABCg-3885 cloning fragment. Lane 1: 1KB Plus DNA

Ladder; Lane 7: 3885 (expected size: 2128bp). Arrow: ABCg-3885 cloning fragment

(29)

4.3.2. Colony PCR of ABCg Entry Clones in D-TOPO Vector

Table 8: Results of cloning ABCg fragments into D-TOPO vector and colony PCR results.

Gene code Amount of plates

Total amount of colonies Amount of colonies tested with colony PCR

Amount of positive colonies

ABCg-3885 2 16 5 5

ABCg-7696 2 17 8 8

Table 8 displays the results after cloning the ABCg fragments into D-TOPO vector and after colony PCR. All the tested entry clones were positive. (See Paragraph 3.4 and 3.9.)

4.3.3. Analysis of ABCg-7696 Entry Clones

Figure 20: ABCg-7696 entry clones in D-TOPO vector put on gel. Lane 1: 1KB Plus DNA Ladder; Lane 2: ABCg-7696 cloning fragment (see Figure 18, expected size: 4302bp); Lane 3-10: ABCg-7696-E1,E2,E3,E4,E5,E6,E7,E8 (expected size: 6882bp (D-TOPO vector plus ABCg-7696)). Arrow: ABCg-7696 cloning fragment.

Figure 21: PCR of ABCg-7696 entry clones in D-TOPO vector. Lane 1: 1KB Plus DNA Ladder; Lane 3-10: ABCg-7696-E6,E1,E2,E7,E5,E8,E4,E3 (expected size: 4302bp (ABCg-7696)).

(30)

Figure 22: Restriction digest of 7696 entry clones in D-TOPO vector. Lane 1: 1KB Plus DNA Ladder; Lane 3-10: ABCg-7696-E3,E4,E8,E5,E7,E2,E1,E6 cut with Not1 and Asc1 (expected sizes: 2580bp (D-TOPO vector) and 4302bp (ABCg-7696)). Figure 20 displays ABCg-7696 cloning fragment and entry clones put on gel. On lane 2 In Figure 20 the band shows a size close to the expected size of the ABCg-7696 cloning fragment. However, bands in lane 3 to 10 show a size that is ~7000bp smaller than the expected size of the ABCg-7696 entry clones. Figure 21 displays the result of a PCR of the ABCg-7696 entry clones on gel. None of the bands showed in figure 21 has a size close to the expected size of the ABCg-7696 cloning fragment. Figure 22 displays a restriction digest of the ABCg-7696 entry clones. None of the bands showed in figure 22 has a size close to the expected sizes of the D-TOPO vector and the ABCg-7696 cloning fragment. (See Paragraph 3.6.)

4.3.4. PCR of ABCg-3885 Destination Clones in pB7GW2 Vector

Table 9: Results of cloning ABCg-3885 fragment into pB7GW2 vector and colony PCR results.

Gene code Amount of plates Total amount of colonies Amount of colonies tested with PCR Amount of positive colonies ABCg-3885 E. coli 2 >350 5 5 ABCg-3885 A. tumefaciens 2 >1000 2 2

Table 9 displays the results after cloning the ABCg-3885 fragment into pB7GW2 vector and after colony PCR. All the tested destination clones were positive. (See Paragraph 3.7, 3.8, 3.9 and 3.11.)

4.3.5 Sequencing Results of ABCg Clones

Table 10: Sequencing results of ABCg entry- and destination clones.

ABCg-3885-E1 (conc increased) with primerR3 0%

ABCg-3885-E5 with primerF3 98%

ABCg-7696-E5 (conc increased) with primerR5 14% ABCg-7696-E7 (conc increased) with primerR5 5%

ABCg-3885-D1 with primerF3 97%

ABCg-3885-D4 (conc increased) with primerF3 96%

Table 10 displays the sequencing results of ABCg entry- and destination clones. Three of the four ABCg-3885 clones share high similary with their respective gene. However, only a small part of the ABCg-7696 entry clones was sequenced. (See Paragraph 3.10.)

Metabolic Engineering of the Feverfew Parthenolide Pathway in N. Benthamiana

Wageningen UR,

Laboratory of Plant

Physiology

Kortooms, Joris

12-06-2014

INTERNSHIP REPORT

M

ETABOLIC

E

NGINEERING

OF

THE

F

EVERFEW

INTERNSHIP REPORT

M

E

F

P

P

N.

Preface

Abstract

Table of Contents

1. Introduction

2. Theoretical Background

2.1. Terpenes

2.2. Sesquiterpene Lactones from

Asteraceae Species

2.3. Metabolic Engineering of Sesquiterpene Lactone Biosynthesis

2.4. Cloning into ImpactTim Entry Vector

2.5. pCR8/GW/TOPO TA Cloning Kit

2.6. pENTR Directional TOPO Cloning Kit

2.7. Gateway Technology: LR Clonase Enzyme Mix II

2.8. Agro-infiltration in N. benthamiana

2.9. Feverfew Candidate Genes

2.10. Project Goal

3. Material and Methods

3.1. PCR of Cloning Fragments

3.2. Cloning into ImpactTim Entry Vector: P450s

3.3. Cloning into TOPO TA Entry Vector: LTP-12309 & LTP-21667

3.4. Cloning into D-TOPO Entry Vector: LTP-14333, LTP-19412, ABCg-3885 &

ABCg-7696

3.5. Restriction Digests: P450s & LTP-21667

3.6. Analysis of ABCg-7696 Entry Clones

3.7. Cloning into Destination Vector with LR Clonase 2: 8272,

P450-8595, P450-9025, LTP-14333, LTP-19412, LTP-21667 & ABCg-3885

3.8. Transformation into A. tumefaciens: 8272, 8595,

P450-9025, LTP-14333, LTP-19412, LTP-21667 & ABCg-3885

3.9. Colony PCR & Plasmid Extraction

3.10. Sequencing of Clones

3.11. PCR of Cloning Fragments from A. tumefaciens Clones: P450-8272,

P450-8595, P450-9025, LTP-14333, LTP-19412, LTP-21667 & ABCg-3885

3.12. Agro-infiltration in N. benthamiana: P450-8595 & P450-9025

3.13. Grinding of Leafs & Extracting Metabolites

3.14. Analysing Leaf Extracts with LC-QTOF-MS

4. Results

4.1. P450s

4.2. LTPs

4.3. ABCGs