University of Groningen Identification, characterisation ... · treatments for uncomplicated...

University of Groningen

Identification, characterisation and expression of early biosynthetic genes from ArtemisiaannuaRydén, Anna-Margareta

IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite fromit. Please check the document version below.

Document VersionPublisher's PDF, also known as Version of record

Publication date:2010

Link to publication in University of Groningen/UMCG research database

Citation for published version (APA):Rydén, A-M. (2010). Identification, characterisation and expression of early biosynthetic genes fromArtemisia annua: for the heterologous biosynthesis of dihydroartemisinic acid as a first step towardsartemisinin production. Groningen: s.n.

CopyrightOther than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of theauthor(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).

Take-down policyIf you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediatelyand investigate your claim.

Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons thenumber of authors shown on this cover page is limited to 10 maximum.

Download date: 28-07-2020

https://www.rug.nl/research/portal/en/publications/identification-characterisation-and-expression-of-early-biosynthetic-genes-from-artemisia-annua(dbefc94b-45bb-405f-a2b6-b3473c08a26f).html




Identification, characterisation and

expression of early biosynthetic genes from

Artemisia annua

for the heterologous biosynthesis of dihydroartemisinic acid

as a first step towards artemisinin production

Anna-Margareta Rydén

2010

RIJKSUNIVERSITEIT GRONINGEN

Identification, characterisation and expression of early biosynthetic genes from Artemisia annua

for the heterologous biosynthesis of dihydroartemisinic acid as a first step towards artemisinin production

Proefschrift

ter verkrijging van het doctoraat in de Wiskunde en Natuurwetenschappen aan de Rijksuniversiteit Groningen

op gezag van de Rector Magnificus, dr. F. Zwarts, in het openbaar te verdedigen op

vrijdag 10 september 2010 om 11.00 uur

door


geboren op 30 juli 1981 te Danderyd, Zweden

Promotores: Prof. dr. O. Kayser Prof. dr. W.J. Quax

Beoordelingscommissie: Prof. dr. P. Brodelius Prof. dr. I.J. van der Klei

Prof. dr. R. Verpoorte

Planning makes the human happy

Anonymous

Paranimfen: Oktavia Hendrawati Magda Czepnik

The studies described in this thesis were performed at:

- the Department of Pharmaceutical Biology, University of Groningen, the Netherlands,

- the Metabolic Regulation Division, Plant Research International, the Netherlands,

- the Antibiotics Laboratory, RIKEN, Japan, - and the Kihara Insitute for Biological Research, Yokohama,

Japan.

Transposition to Prepress by: Grafimedia, University of Groningen, Groningen, the Netherlands. Printed by: Grafimedia, University of Groningen, Groningen, the Netherlands.

ISBN printed version: 978-90-367-4493-5 ISBN digital version: 978-90-367-4494-2

Contents

Chapter 1............................................................................................ 9 Introduction and scope of the thesis .......................................... 9

Chapter 2.......................................................................................... 24 Chemistry, biosynthesis and biological activity of artemisinin and related natural peroxides................................................... 24

Chapter 3.......................................................................................... 75 Molecular cloning and characterization of a broad substrate terpenoid oxidoreductase from Artemisia annua .................... 75

Chapter 4........................................................................................ 103 The molecular cloning of dihydroartemisinic aldehyde reductase and its implication in artemisinin biosynthesis in Artemisia annua ....................................................................... 103

Chapter 5........................................................................................ 125 Oxygen dependent biosynthesis of artemisinin precursors in yeast .......................................................................................... 125

Chapter 6........................................................................................ 163 Role of peroxygenase in artemisinin production .................. 163

Chapter 7........................................................................................ 185 Summary, general discussion and future perspectives ........ 185

Chapter 8........................................................................................ 193 Nederlandse samenvatting ...................................................... 193

Acknowledgements........................................................................ 201 Publication list ............................................................................... 207 Curriculum vitae ............................................................................ 211

Chapter 1

Introduction and scope of the thesis

Muntendam, R.1, Melillo, E.1, Ryden, A.M.1, and Kayser, O1.

1Department of Pharmaceutical Biology, GUIDE, University of Groningen, 9713 AV Groningen, the Netherlands

Free from: Perspectives and limits of engineering the isoprenoid metabolism in heterologous hosts (2009), Appl Microbiol Biotechnol. vol. 84, Issue 6, p.1003-19.

Chapter 1

10

Terpenoids and their role in nutrition and medicine

Terpenoids belong to the largest class of natural products with important medicinal and industrial properties, and today approximately 25,000 terpenoid structures have been elucidated and isolated from plants, microorganisms, insects and various marine organisms (Gershenzon et al., 2007). Despite the enormous structural complexity, terpenoids are constructed from two basic isoprene building blocks, isopentenyl pyrophosphate and dimethylallyl pyrophosphate, which can be coupled in nearly any way as defined by the enzymatic setup in a species.

Terpenoids serve a number of different functions e. g., steroids for membrane fluidity and hormone signaling, carotenoids for photosynthesis in plants and as antioxidant agents (Howitt et al., 2006), quinones for electron transport (Berry, 2002) and dolichol for protein glycosylation (Lehrman, 2007). Especially from plants and marine invertebrates have many terpenoids with significant pharmaceutical importance been identified. Indeed, the terpenoids encompass a vast number of bioactive compounds like artemisinin, paclitaxel and menthol; most of these molecules are used in pharmaceutical, cosmetic or food industries. Compounds such as limonene, menthol and carotenoids are available in discrete quantities in Nature, and hence they can easily be extracted and purified. However, terpenoids such as the antiplasmodial artemisinin and the anti-cancer chemotherapy drug paclitaxel are produced by one dedicated species respectively and furthermore in small quantities. Such limitations cause supply problems. This has sparked an interest in using biotechnology to solve the situation.

Metabolic engineering of rare and complex compounds such as artemisinin or paclitaxel is a new concept in synthetic biology. The strategy is considered a novel avenue for the production of rare and high-cost natural products. The underlying idea encompasses transfer of a complete plant terpenoid pathway to microorganisms to


11

optimize the biosynthetic production rate by fermentation. However, pathways in both bacteria and fungi meant to be used as target hosts are regulated through complex networks. Furthermore, the regulation of the terpenoid pathway is far from simple and control mechanisms are rather unknown. In chapter 5, the use of yeast as a heterologous host for production of artemisinin intermediates is described.

The keen interest in artemisinin, a sesquiterpene lactone which is extracted from the sweet wormwood plant Artemisia annua (A.annua), emanates from the potent antimalarial property of the molecule, and its lack of side effects (Li et al., 2010). Furthermore, artemisinin has been shown to reduce cancer by inducing apoptosis, disrupting cell migration and cell cycle arrest as well as inhibiting angiogenesis (Firestone et al., 2009). Moreover, due to development of resistance against traditional antimalarial medicines such as chloroquine, artemisinin combination therapies are left as the sole remaining treatment of the disease (Eastman et al., 2009). Artemisinin combination therapies are therefore the first-line treatments for uncomplicated Plasmodium falciparum (P.falciparium) malaria in most malaria-endemic countries. With 150-300 children lost to malaria each hour and approxiamtely a toll of 2 million deaths a year, malaria has been put forward as one of the most urgent catastrophies by the World Health Organization (Breman, 2009). This has been followed up by an intense research in how artemisinin can be synthesized, most efficiently isolated from the plant, in how the yield in planta can be increased and the latetst approach aimed at heterlogous production of artemisinin (chapter2).

Metabolic engineering of the terpenoid pathway

The concept of metabolic engineering

Metabolic engineering is considered as one of the major concepts in biotechnology. At the moment, genetic engineering allows the

DocWorks

Distance Measurement

Distance: 17 mm

DocWorks


Distance: 24,03 mm

DocWorks


Distance: 0 mm

Chapter 1

12

transfer of a biosynthetic pathway to any selected host. The isoprenoid pathway is an example on how biosynthetic relevant genes can be reassembled from different biological sources (e.g. plants, bacteria and fungi) in a heterologous microorganism. Engineering of plant terpenoids into microbial hosts has been focused mainly on isoprenoid derived compounds such as carotenoids (Schmidt-Dannert et al., 2000), artemisinin and paclitaxel (Newman, et al., 2006; Withers et al., 2007). The production of carotenoids and artemisinin demonstrates that complex natural products can be produced by microbial fermentation with yields approaching commercial relevance. Development of a microbial production platform for the biosynthesis of complex terpenoids offers large-scale and cost-effective industrial production via fermentation, which is independent from climate (too dry, too wet) and cultivation risks (pest controls, weeds, soil conditions). Well characterized and genetically fairly easy to manipulate heterologous hosts such as Escherichia coli (E. coli) and Saccharomyces cerevisiae (S. cerevisiae) allow very specific engineering of biosynthetic pathways for increased yields and generating novel compounds. Although extensive engineering is frequently required to successfully reconstitute biosynthetic pathways in these hosts, the potential to manipulate the biosynthesis is a significant advantage when compared to engineering in poorly characterized host strains or non microbial systems.

Selecting microorganisms for engineering terpenoid pathways

Commonly the host E. coli and S. cerevisiae are favored in metabolic engineering, but other hosts can be used as well. Both laboratory domestic organisms share the advantages of having been studied over decades and the elucidation of their genome, transcriptome and metabolome are nearly complete. However, in silico models of these organisms do not exist (Feist et al., 2009). In theory, the possibility exists to adapt every host for the production of natural compounds, but as various organisms exhibit different genetic properties, they


13

also react distinctively to genetic manipulations and to the introduced recombinant genes. Although nearly all organisms are able to produce terpenoid structures (Goldstein et al., 1990; Hunter, 2007), their strategy and diversity varies dramatically between species. Synthetic biology provides the tools to overcome these problems and to explore the rich variety of enzymes by which nature is capable of creating all imaginable structures. This can be exploited by metabolic engineers: by selecting the right host for recombinant expression, it is possible to increase production levels by minimal adaption of the genetic system. This strategy was investigated as described in chapter 5.

Engineering the secondary terpenoid metabolism

In contrast to the primary metabolism, no standard biological components exist which would fit for all different secondary pathways. Secondary metabolite routes are per definition species specific and therefore we have to expect a high diversity in structures between organisms in the terpenoid biosynthesis (Fischbach et al., 2007). Artemisinin is an excellent example since A. annua produces low amounts of artemisinin, while its relatives Artemisia absinthium(Van Nieuwerburgh, et al., 2006) and Artemisia afra (Van der Kooy et al., 2008) synthesize no artemisinin at all. The properties of many terpenoids make them interesting for recombinant production, although many pathways are yet too complex or enzymes unknown. This thesis reports on the isolation and functional characterization of enzymes from the artemisinin biosynthetic pathway (chapter 3-6).

Case study artemisinin

The activity of artemisinin against the malaria parasite P.falciparium has been well established. In addition, it has been proven that artemisinin acts by selective inhibition of the sarco/endoplasmic reticulum Ca2+ ATPase (SERCA). The same study showed that artemisinin contributes to a change in membranous structures in the

Chapter 1

14

cytoplasm of parasites inside erythrocytes (Eckstein-Ludwig, et al., 2003). Examples of derivatives of artemisinin include artemether and artemisone, which are on the market or currently under development (Haynes, et al., 2006). An overview of artemisinin derivatives is described in chapter 2.

The low production yield of artemisinin from A. annua (0.01-1 % of dry weight) has led to difficulties in managing the demand, while offering an acceptable price for most patients (Liu et al., 2006; Mutabingwa, 2005). The wish to improve the overall supply of artemisinin at a reduced market price has encouraged interest in molecular biology and biochemistry of artemisinin biosynthesis (figure 1) (Bertea, et al., 2005; Bouwmeester, et al., 1999; Covello et al., 2007; Mercke et al., 2000; Ro, et al., 2006). The biosynthesis starts with the cyclization of farnesyl diphosphate to amorpha-4,11-diene by amorpha-4,11-diene synthase (Amds) (Kim, et al., 2006; Picaud, et al., 2006; Wallaart et al., 2001). Various terpene synthases are known to convert the terpenoid backbone into pathway specific structures, however some of them are less specific and can produce more than one structure (Christianson, 2008). In the first step a divalent metal, generally magnesium, is used to stabilize the pyrophosphate group in similar terpenoid backbones, such as geranyl diphosphate or farnesyl diphosphate. After this first committed enzymatic step towards secondary relevant terpenoids, the pyrophosphate is lost and more species specific enzymes (e.g. cytochromes) create the broad diversity of structures (Covello, et al., 2007; Croteau et al., 2005).

Subsequent enzymatic steps in the biosynthetic pathway of artemisinin can possibly be performed in different orders. The main likely conversion scheme in planta for production of artemisinin include the oxidation of amorpha-4,11-diene to artemisinic alcohol followed by a further oxidation at C12 from alcohol to aldehyde by the cytochrome P450 Cyp71av1. In the next step, the carbon double


15

Figure 1. Biosynthesis of artemisinin in A. annua. The pathway is a summary based on chapters 2-5 as well as the literature mentioned therein. AMDS – Amorpha-4,11-diene synthase, Cyp71av1 – Cytochrome P450 71av1, Dbr2 – Double bond reductase 2, FPP – Farnesyl diphosphate, Red1 – Reductase 1.

PPO 111

FPP

PPOH

7

32

+

Bisabolyl carbocation

+

+

Amorpha-4,11-diene

1

15

10

614

13

11

AMDS AMDS

AMDS AMDS

AMDS

OH

H

Artemisinic alcohol

Cyp71av1

O

H

Artemisinic aldehyde

Cyp71av1

O

OH

Artemisinic acid

Cyp71av1

OH

H

Dihydroartemisinic alcohol

Cyp71av1

O

H

Dihydroartemisinic aldehyde

Cyp71av1

O

OH

Dihydroartemisinic acid

Dbr2

Red1

Artemisinin

Artemisinin

Chapter 1

16

bond at C11-C13 is reduced forming dihydroartemisinic aldehyde by the artemisinic aldehyde reductase (Dbr2), followed by conversion of dihydroartemisinic aldehyde to dihydroartemisinic acid by Cyp71av1. The formation of the 1,2,4-trioxane moiety is not fully understood and is suggested to occur patially in a non-enzymatic manner (Covello, et al., 2007; Ro, et al., 2006; Rydén et al., 2007). Identification of these enzymes is essential for the production of artemisinin in heterologous systems. These issues are further discussed in chapter 2.

The value of synthetic biology was proven when the two genes Amds and Cyp71av1 were expressed in yeast. The transgenic yeast produced more artemisinic acid in contrast to relative plant biomass (4.5% dry weight in yeast compared to 1.9% artemisinic acid and 0.16% artemisinin in A. annua) and in a shorter time period (4–5 days for yeast versus several months for A. annua) (Ro, et al., 2006). When these two enzymes were co-expressed in yeast together with Dbr2, dihydroartemisinic acid was produced at a level of 15.7 (+/-1.4) mg/L culture and the related artemisinic acid was produced at levels of 11.8 mg/L (Zhang, et al., 2008). The production level of artemisinic acid was reduced compared to the mutant without the artemisinic aldehyde reductase (29.4 mg/L), due to production of dihydroartemisinic acid. This is preferred because dihydroartemisinic acid is chemically converted to artemisinin both easier and cheaper than artemisinic acid.

Perspectives

In the future, metabolic engineering will provide a third alternative to plant cell culture and chemistry to cover the high demand of the pharmaceutical drug artemisinin for a reasonable price. Pathway engineering will provide more benefits compared to traditional approaches: stereo-chemically complex natural products such as artemisinin can hardly be synthesized by conventional synthetic


17

chemistry. This approach still needs to be adapted to minimize unwanted side products in the synthesis. A biosynthetic production can be performed at room temperature, under buffer media conditions and under atmospheric pressure. This procedure is termed white biotechnology and it supports the transition from classical “dirty” organic chemistry. It is gaining support worldwide, where companies strongly associated with chemistry are investing in alternatives based on evolved proteins to improve industrial processes and create novel compounds (Ernst&Young, 2007). Upcoming white biotechnology will benefit from synthetic biology and metabolic engineering and change the way to synthesize even small terpenoids. A summary of the results presented in this thesis along with future perspectives is presented in chapter 7 and as a Dutch summary in chapter 8.

References

Berry, S. (2002). The chemical basis of membrane bioenergetics. JMol Evol, 54(5), 595-613.

Bertea, C. M., Freije, J. R., van der Woude, H., Verstappen, F. W., Perk, L., Marquez, V., et al. (2005). Identification of intermediates and enzymes involved in the early steps of artemisinin biosynthesis in Artemisia annua. Planta Med, 71(1), 40-47.

Bouwmeester, H. J., Wallaart, T. E., Janssen, M. H., van Loo, B., Jansen, B. J., Posthumus, M. A., et al. (1999). Amorpha-4,11-diene synthase catalyses the first probable step in artemisinin biosynthesis. Phytochemistry, 52(5), 843-854.

Breman, J. G. (2009). Eradicating malaria. Sci Prog, 92(Pt 1), 1-38. Christianson, D. W. (2008). Unearthing the roots of the terpenome.

Curr Opin Chem Biol, 12(2), 141-150. Covello, P. S., Teoh, K. H., Polichuk, D. R., Reed, D. W., & Nowak,

G. (2007). Functional genomics and the biosynthesis of artemisinin. Phytochemistry, 68(14), 1864-1871.

Chapter 1

18

Croteau, R. B., Davis, E. M., Ringer, K. L., & Wildung, M. R. (2005). (-)-Menthol biosynthesis and molecular genetics. Naturwissenschaften, 92(12), 562-577.

Eastman, R. T., & Fidock, D. A. (2009). Artemisinin-based combination therapies: a vital tool in efforts to eliminate malaria. Nat Rev Microbiol, 7(12), 864-874.

Eckstein-Ludwig, U., Webb, R. J., Van Goethem, I. D., East, J. M., Lee, A. G., Kimura, M., et al. (2003). Artemisinins target the SERCA of Plasmodium falciparum. Nature, 424(6951), 957-961.

Ernst&Young. (2007). Sustained progress, the European perspective. Beyond borders: The global biotechnology report 2007, 44-47.

Feist, A. M., Herrgard, M. J., Thiele, I., Reed, J. L., & Palsson, B. O. (2009). Reconstruction of biochemical networks in microorganisms. Nat Rev Microbiol, 7(2), 129-143.

Firestone, G. L., & Sundar, S. N. (2009). Anticancer activities of artemisinin and its bioactive derivatives. Expert Rev Mol Med, 11, e32.

Fischbach, M. A., & Clardy, J. (2007). One pathway, many products. Nat Chem Biol, 3(7), 353-355.

Gershenzon, J., & Dudareva, N. (2007). The function of terpene natural products in the natural world. Nat Chem Biol, 3(7), 408-414.

Goldstein, J. L., & Brown, M. S. (1990). Regulation of the mevalonate pathway. Nature, 343(6257), 425-430.

Haynes, R. K., Fugmann, B., Stetter, J., Rieckmann, K., Heilmann, H. D., Chan, H. W., et al. (2006). Artemisone--a highly active antimalarial drug of the artemisinin class. AngewChem Int Ed Engl, 45(13), 2082-2088.

Howitt, C. A., & Pogson, B. J. (2006). Carotenoid accumulation and function in seeds and non-green tissues. Plant Cell Environ, 29(3), 435-445.

Hunter, W. N. (2007). The non-mevalonate pathway of isoprenoid precursor biosynthesis. J Biol Chem, 282(30), 21573-21577.


19

Kim, S. H., Heo, K., Chang, Y. J., Park, S. H., Rhee, S. K., & Kim, S. U. (2006). Cyclization mechanism of amorpha-4,11-diene synthase, a key enzyme in artemisinin biosynthesis. J Nat Prod, 69(5), 758-762.

Lehrman, M. A. (2007). Teaching dolichol-linked oligosaccharides more tricks with alternatives to metabolic radiolabeling. Glycobiology, 17(8), 75R-85R.

Li, Q., & Weina, P. J. (2010). Severe embryotoxicity of artemisinin derivatives in experimental animals, but possibly safe in pregnant women. Molecules, 15(1), 40-57.

Liu, C., Zhao, Y., & Wang, Y. (2006). Artemisinin: current state and perspectives for biotechnological production of an antimalarial drug. Appl Microbiol Biotechnol, 72(1), 11-20.

Mercke, P., Bengtsson, M., Bouwmeester, H. J., Posthumus, M. A., & Brodelius, P. E. (2000). Molecular cloning, expression, and characterization of amorpha-4,11-diene synthase, a key enzyme of artemisinin biosynthesis in Artemisia annua L. Arch Biochem Biophys, 381(2), 173-180.

Mutabingwa, T. K. (2005). Artemisinin-based combination therapies (ACTs): best hope for malaria treatment but inaccessible to the needy! Acta Trop, 95(3), 305-315.

Newman, J. D., Marshall, J., Chang, M., Nowroozi, F., Paradise, E., Pitera, D., et al. (2006). High-level production of amorpha-4,11-diene in a two-phase partitioning bioreactor of metabolically engineered Escherichia coli. Biotechnol Bioeng, 95(4), 684-691.

Picaud, S., Mercke, P., He, X., Sterner, O., Brodelius, M., Cane, D. E., et al. (2006). Amorpha-4,11-diene synthase: mechanism and stereochemistry of the enzymatic cyclization of farnesyl diphosphate. Arch Biochem Biophys, 448(1-2), 150-155.

Ro, D. K., Paradise, E. M., Ouellet, M., Fisher, K. J., Newman, K. L., Ndungu, J. M., et al. (2006). Production of the antimalarial drug precursor artemisinic acid in engineered yeast. Nature, 440(7086), 940-943.

Chapter 1

20

Rydén, A.-M., & Kayser, O. (2007). Chemistry, Biosynthesis and Biological Activity of Artemisinin and Related Natural Peroxides Bioactive Heterocycles III (pp. 1).

Schmidt-Dannert, C., Umeno, D., & Arnold, F. H. (2000). Molecular breeding of carotenoid biosynthetic pathways. NatBiotechnol, 18(7), 750-753.

Van der Kooy, F., Verpoorte, R., & Marion Meyer, J. J. (2008). Metabolomic quality control of claimed anti-malarial Artemisia afra herbal remedy and A. afra and A. annua plant extracts. South African Journal of Botany, 74(2), 186.

Van Nieuwerburgh, F. C. W., Vande Casteele, S. R. F., Maes, L., Goossens, A., Inzé, D., Van Bocxlaer, J., et al. (2006). Quantitation of artemisinin and its biosynthetic precursors in Artemisia annua L. by high performance liquid chromatography-electrospray quadrupole time-of-flight tandem mass spectrometry. Journal of Chromatography A, 1118(2), 180.

Wallaart, T. E., Bouwmeester, H. J., Hille, J., Poppinga, L., & Maijers, N. C. (2001). Amorpha-4,11-diene synthase: cloning and functional expression of a key enzyme in the biosynthetic pathway of the novel antimalarial drug artemisinin. Planta, 212(3), 460-465.

Withers, S. T., & Keasling, J. D. (2007). Biosynthesis and engineering of isoprenoid small molecules. Appl Microbiol Biotechnol, 73(5), 980-990.

Zhang, Y., Teoh, K. H., Reed, D. W., Maes, L., Goossens, A., Olson, D. J., et al. (2008). The molecular cloning of artemisinic aldehyde Delta11(13) reductase and its role in glandular trichome-dependent biosynthesis of artemisinin in Artemisia annua. J Biol Chem, 283(31), 21501-21508.

Chapter 2

Chemistry, biosynthesis and biological activity of

artemisinin and related natural peroxides

Anna-Margareta Rydén1 and Oliver Kayser1

1Department of Pharmaceutical Biology, GUIDE, University of Groningen, 9713 AV Groningen, the Netherlands

Adapted from: Bioactive Heterocycles III, series: Topics in Heterocyclic Chemistry (2007), vol. 9, p. 1-32

Chapter 2

22

Abstract

Artemisinin is a heterocyclic natural product and belongs to the natural product class of sesquiterpenoids with an unusual 1,2,4 trioxane substructure. Artemisinin is one of the most potent antimalarial drugs available. It serves as a lead compound in the drug development process to identify new chemical derivatives with optimized antimalarial activity and improved bioavailability. In this review we report about the latest status of research on chemical and physical properties of the drug and its derivatives. We describe new strategies to produce artemisinin on a biotechnological level in heterologous hosts and in plant cell cultures. We also summarize recent reports on its pharmacokinetic profile. Furthermore we describe attempts to develop drug delivery systems to overcome bioavailability problems and to target the drug to Plasmodiuminfected erythrocytes as main target cells.

Artemisinin – chemistry and biology

23

Chemistry

For thousands of years Chinese herbalists treated fever with a decoction of the plant called "qinghao", Artemisia annua, "sweetwormwood" or "annual wormwood", belonging to the family of Asteraceae. In the 1960s a program of the People Republic of China re-examined traditional herbal remedies on a rational scientific basis including the local qinghao plant. Early efforts to isolate the active principle were disappointing. In 1971 Chinese scientists followed an uncommon extraction route using diethyl ether at low temperatures obtaining an extract with a compound that was highly active in vivoagainst Plasmodium berghei in infected mice. The active ingredient was febrifuge, structurally elucidated in 1972, called mostly in China "qinghaosu", or "arteannuin" and in the west "artemisinin". Artemisinin, a sesquiterpene lactone, contains a peroxide group unlike most other antimalarials. It was also named artemisinine, but following IUPAC nomenclature a final "e" would suggest that it was a nitrogen-containing compound which is misleading and therefore that name is not favored today.

Artemisinin and its antimalarial derivatives belong to the chemical class of unusual 1,2,4-trioxanes. Artemisinin is poorly soluble in water and decomposes in other protic solvents, probably by opening of the lactone ring. It is soluble in most aprotic solvents and is unaffected by them at temperatures up to 150 °C and shows a remarkable thermal stability. This section will focus on biological and pharmaceutical aspects; synthetic routes to improve antimalarial activity and synthetic production of artemisinin derivatives with different substitution patterns are reviewed elsewhere (Woerdenbag, et al., 1994; Ziffer et al., 1997). Most of the chemical modifications were conducted to modify the lactone function of artemisinin to a lactol. In general alkylation, or a mixture of dihydroartemisinin epimers in the presence of an acidic catalyst, gave products with predominantly -orientation, whereas acylation in alkaline medium preferentially yields -orientation (figure 1). Artemether (figure 1.2)

Chapter 2

24

as the active ingredient of Paluther is prepared by treating a methanol solution of dihydroartemisinin with boron trifluoride-etherate, yielding both epimers. The main goal was to obtain derivatives that show a higher stability when dissolved in oils to enable parenteral use. The -epimer is slightly more active (EC50 = 1.02 mg/kg b.w.) than the -epimer (EC50 = 1.42 mg/kg) and artemisinin itself (EC50 = 6.2 mg/kg) (CCRG, 1982). Synthesis of derivatives with enhanced water solubility has been less successful. Sodium artesunate, Arsumax (1.5) has been introduced in clinics, is well tolerated and less toxic than artemisinin.

O

O

CH3

O

O

OR

CH3

H

CH3

Figure 1. Artemisinin and its derivatives. (1.1) dihydroartemisin; R = H ( + ß), (1.2) artemether; R = CH3 (ß), (1.3) arteether; R = CH2CH3 (ß), (1.4) artelinate; R = CH2C6H4COONa (ß), (1.5) artesunate; R = COCH2CH2COONa ( ).

Trioxane and peroxides in nature

Besides artemisinin more than 150 natural peroxides are known in Nature. The presence of the typical peroxide functions is not related to one natural product group and occurs as cyclic and acyclic peroxides in terpenoids, polyketides, phenolics and also alkaloids. Most stable are cyclic peroxides even under harsh conditions and artemisinin is a nice example for that. Artemisinin can be boiled or treated with sodium borohydride as base without degradation of the


25

peroxide function. In contrast acyclic peroxides are rather unstable, form hydrogen peroxides and are easily broken by metals or bases.

Most of the natural peroxides have been isolated from plants and marine organisms, and terpenoids have attracted most interest because of the structural diversity they cover. In an excellent review from Jung et al.(2003), an overview is given and it should be stressed out that Scapania undulata, which is a bryophyte in the Nothern parts of Europe, biosynthesize amorphane like natural products with a cyclic peroxide (figure 2.1) structurally related to the well known artemisinin. There is less information about the biological activity of natural peroxides from plant origins, but some reports indicate the use against helminth infections, rheumatic diseases and antimicrobial activity. Natural cyclic peroxides from marine source (figure 2) have been tested for a broad range of activities including antiviral (Aikupikoxide A), antimalarial, antimicrobial activity and cytotoxicity (figure 2.2). A second important natural product group is polyketides and it is interesting that all of the isolated polyketide-derived peroxides are from marine sources. Due to the high flexibility in the carbon chain and presence of hydroxy substituents, a high chemical diversity can be documented ranging from simple and short peroxides like haterumdioins in Japanese sponge Plaktoris lita to more complex structures with long chain derivatives like peroxyacarnoic acids from the sponge Acarnus bicladotylota (figure 2.3). Most of the polyketide-derived peroxides show a high cytotoxic activity and moderate activity against microorganisms.

As expected, due to chemical instability the number of acyclic peroxides is lower. Most of them occur as plant derived products, but also in soft corals like Clavularia inflata, hydroperoxides with potent cytotoxicity exist. Interestingly the bioactivity disappeared when the hydroperoxide function was deleted. It must be noted that most of natural hydroperoxides in plants are found in the group of saponins from Panax ginseng or Ficus microcarpa which are used in the ethnomedicine in South East Asia.

Chapter 2

26

Figure 2. Natural peroxides. (2.1) Terpene type peroxide from Scapania undulate, (2.2) Aikupikoxide-D from Diacarnus erythraenus, (2.3) Peroxyacarnoic acid A from Acarnusbicladotyloata.

Biosynthesis

Biosynthesis in Artemisia annua

Biochemistry

There are two pathways employed in plants for production of isoprenoids, the 1-deoxy-D-xylulose 5-phosphate pathway (DXP) localized in the plastid and the mevalonate pathway which is precent in the cytosol (figure 3) (Yan Liu et al., 2005). These pathways are normally used to produce different sets of isoprenoids; sesquiterpenoids, sterols and triterpenoids among others being reserved for the mevalonate pathway, while the diterpenes and

(2.1)

OO

OH

OO

OO

O

OH

HO

OO

O

(2.2)

(2.3)


27

R

H

H

AACT

2 Acetyl-CoA Acetoacetyl-CoA Glyceraldehyde 3-P Pyruvate

HMGS DXS

HMG-CoA DXP

HMGR DXR

MVA MEP

MK CMS

MVAP CDP-ME

MPK CMK

MPP CDP-MEP

MDD MCS

DMAPP IPP cMEPP HDS

FPPS HMBPP Squalene Polyprenols FPP Prenylation Sesquiterpenes AMDS

IDS IDS Amorpha-4,11-diene IPP DMAPP

CYP71AV1

IPPi

GPP

GPPS

Monoterpenes

Diterpenes

Carotenoids

Quinones

Gibberellins

R = CH2OH, artemisinic alcohol

R = CHO, artemisinic aldehyde

R = COOH, artemisinic acid

Dihydro foms when exocyclic double

bond is reduced

H

HCOOH


H

COOH

OOH


hydroperoxide O

O

OO

O

O O

O

Arteannuin B

Artemisinin

1O2

3O2

Cytosol

Endoplasmatic reticulum

Plastid

Peroxidase

Peroxidase

IPPi

CH3

H

H

Chapter 2

28

Figure 3. Isoprenoid biosynthetic pathways in plant cells. In the cytosol the mevalonate pathway is represented, in the plastid the MEP pathway. Biosynthesis of artemisinin is depicted in detail. Long dash arrow depicts transport. Dash punctured arrow depicts unknown or putative enzymatic function. Single arrow depicts single reaction step. Multiple arrows depict several reaction steps. Shortenings of substrates: CDP-ME, 4-(Cytidine 5’-diphospho)-2-C-methyl-D-erythritol; CDP-MEP, 4-(Cytidine 5’-diphospho)-2-C-methyl-D-erythritol 2-phosphate; cMEPP, 2-C-Methyl-D-erythritol 2,4-cyclodiphosphate; DMAPP, Dimethylallyl diphosphate; DXP, 1-Deoxy-D-xylulose 5-phosphate; FPP, Farnesyl diphosphate; GPP, Geranyl diphosphate; HMBPP, 1-Hydroxy-2-methyl-2-(E)-butenyl 4-diphosphate; HMG-CoA, 3S-Hydroxy-3-methylglutaryl-CoA; IPP, Isopentenyl diphosphate; MEP, 2-C-Methy-D-erythritol 4-phosphate; MPP, Mevalonate diphosphate; MVA, 3R-Mevalonic acid; MVAP, Mevalonic acid-5-phosphate. Shortenings of enzymes: AACT, Acetoacetyl-coenzyme A (CoA) thiolase; AMDS, Amorpha-4,11-diene synthase; CMK, 4-(Cytidine 5’-diphospho)-2-C-methyl-D-erythritol kinase; CMS, 2-C-Methyl-D-erythritol 4-phosphate cytidyl transferase; CYP71AV1, Cytochrome P450 71AV1; DXR, 1-deoxy-D-xylulose 5-phosphate reductoisomerase; DXS, 1-deoxy-D-xyluose 5-phosphate synthase; FPPS, Farnesyl diphosphate synthase; GPPS, Geranyl diphosphate synthase; HDS, 1-Hydroxy-2-methyl-2-(E)-butenyl 4-diphosphate synthase; HMGR, 3-hydroxy-3-methylglutaryl CoA reductase; HMGS, 3-hydroxy-3-methylglutaryl CoA synthase; IPPi, Isopentenyl diphosphate isomerase; MCS, 2- C-Methyl-D-erythritol 2,4-cyclodiphosphate synthase; MDD, mevalonate diphosphate decarboxylase; MK, mevalonate kinase; MPK, mevalonate-5-phosphate kinase.


29

monoterpenes are produced by the DXP pathway. However, there are recent evidences that the pathways to some extent crosstalk at the isopentenyl diphosphate (IPP) level (Yan Liu, et al., 2005).

The first committed step in the biosynthetic pathway of artemisinin is the cyclization of the general mevalonate pathway sesquiterpenoid precursor farnesyl diphosphate (FPP) into (1S, 6R, 7R, 10R)-amorpha-4,11-diene by amorpha-4,11-diene synthase (AMDS) (figure 4) (Chang et al., 2000; Mercke et al., 2000; Wallaart et al., 2001). The crystal structure of this sesquiterpene synthase is not known which is the situation for all but one plant sesquiterpene synthase: the 5-epi-aristolochene synthase from tobacco (Starks et al., 1997). In contrast, the mechanism behind the cyclization of FPP into amorpha-4,11-diene has been proven by Picaud et al. (2006) and Kim et al. (2006) through the use of deuterium labeled FPP (figure 4). Differing from the bicyclic sesquiterpene cyclases -cadinene synthase from cotton (Benedict et al., 2001) and pentalene synthase (Cane et al., 2003) which produce a germacrene cation as the first cyclic intermediate, Amds produces a bisabolyl cation. FPP is ionized and the paired diphosphate anion (OPP) is transferred to C3 giving (3R)-nerolidyl diphosphate. This intermediate allows rotation around the C2-C3 bond to generate a cisoid form. The cisoid form brings C1 in close proximity to C6 allowing a bond formation between these two carbon atoms thus resulting in the first ring closure and a bisabolyl cation. The formed cation is in equilibrium with its deprotonized uncharged form, which is interesting because it implies a solvent proton acceptor and stands in contrast to studies discussing properties of the active site of a investigated trichodiene synthase (Rynkiewicz et al., 2001).

Chapter 2

30

a

PPO 111

FPP

PPOH

7

32

+

Bisabolyl carbocation

+

+

Amorpha-4,11-diene

1

15

10

614

13

11

b

PPO 1 11

PPO

H

+

Helminthogermacradienylcarbocation

Figure 4. Commitment of FPP to amorpha-4,11-diene. a. Cyclization of FPP to amorpha-4,11-diene by Amds as

described by Kim et al. (2006) and Picaud et al. (2006). b. Cyclization of FPP to helmonthogermacradienyl

carbocation.


31

Rynkiewicz and Cane came to the conclusion that the active site is completely devoid of any solvent molecule which would prematurely quench the reaction. In a second report from the group Vedula et al. (2005), the authors draw the conclusion from their results that terpene cyclization reactions in general are governed by kinetic rather than thermodynamic rules in the step leading to formation of the carbocation. In the bisabolyl cation, an intermediate in the reaction towards amorpha-4,11-diene, a 1,3 hydride shift to C7 occurs leaving a cation with a positive charge at C1 (FPP numbering).Through a nucleophilic attack on C1 by the double bond C10-C11 the second ring closes to give an amorphane cation. Deprotonation on C12 or C13 (amorphadiene numbering) gives amorpha-4,11-diene.

The three-dimensional structures of three non-plant sesquiterpene synthases reveals a single domain composed entirely of -helices and loops despite the low homology on amino acid sequence level (Caruthers et al., 2000; Lesburg et al., 1997; Rynkiewicz, et al., 2001). The secondary elements of 5-epi-aristolochene synthase, a plant sesquiterpene synthase, conforms to this pattern with the exception that it contains two domains solely composed of -helices and loops. It is reasonable, but still a matter of debate, to extrapolate these data to the case of amorpha-4,11-diene synthase which probably will only display -helices and loops once the crystal structure has been solved.

A further element shared by all sesquiterpene synthases is the need for a divalent metal ion as cofactor. The metal ion is essential for substrate binding but also for product specificity. The metal ions stabilize the negatively charged pyrophosphate group of farnesyl diphosphate as illustrated by the crystal structure of 5-epi-aristolochene synthase (Starks, et al., 1997). The highly conserved sequence (I, L, V)DDxxD(E) serves to bind the metal ions in all known terpene and prenyl synthases (figure 5) (Colby et al., 1993; Desjardins et al., 1993; Hohn et al., 1989; Math et al., 1992; Proctor

Chapter 2

32

et al., 1993). A further interesting property among terpene synthases is that the active sites are enriched in relatively inert amino acids, thus it is the shape and dynamic of the active site that determines catalytic specificity (Greenhagen et al., 2006).

Figure 5. Computerized 3D-structure of amorpha-4,11-diene.Residues marked with space fill reprecentation belong to the conserved metal ion binding amino acid sequence IDDxxDD. We kindly thank Wolfgang Brandt, Leibniz Institute of Plant Biochemistry Halle/Germany for his 3D model of amorphadiene synthase (Amds).

Picaud et al. (2005) purified recombinant Amds and determined its pH optimum to 6.5. Several sesquiterpene synthases show maximum activity in this range, examples are tobacco 5-epi-aristolochene synthase (Bouwmeester et al., 2002; Vogeli et al., 1990), germacrene A synthase from chickory (Bouwmeester, et al., 2002) and nerolidol synthase from maize (Schnee et al., 2002). Terpenoid synthases are, however, not restricted to a pH optimum in this range. Intriguing examples are the two (+)- -cadinene synthase variants from cotton


33

which exhibit maximum activity at pH 8.7 and 7-7.5 respectively (Chen et al., 1996) and 8-epi-cedrol synthase from A. annua (Merckeet al., 1999) with pH optimum around 8.5-9.0. The authors further investigated the metal ion required as cofactor for Amds as well as substrate specificity. The kinetics studies revealed kcat/Km values of 2.1x10-3 μM-1 s-1 for conversion of FPP at the pH optimum 6.5 with Mg2+ or Co2+ ions as cofactors and a slightly lower value of 1.9x10-3

μM-1 s-1 with Mn2+ as a cofactor. These very low efficiencies are common to several sesquiterpene synthases but there are substantial differences reported. The synthase reached a kcat/Km value of 9.7x10-

3 μM-1 s-1 for conversion of FPP at pH 9.5 using Mg2+ as a metal ion cofactor. This increase in efficiency is interesting and shows the broad window in which the enzyme can work, something that may prove to be industrially useable but which physiologically does not have a meaning in the plant. The increase in efficiency is not linear as the maximum activity of Amds is around pH 6.5-7.0 with a minimum at pH 7.5. The established pH optimum of 6.5 is in line with the range established for Amds isolated from A. annua leaves (Bouwmeester et al., 1999). Amds did not show any relevant activity in the presence of Ni2+, Cu2+ or Zn2+. In presence of Mn2+ as cofactor, Amds is capable of using geranyldiphosphate (GPP) as substrate although with very low efficiency (4.2 x10-5 μM-1 s-1 at pH 6.5). Using Mn2+ as a cofactor also increased the product specificity of Amds to ~ 90% amorpha-4,11-diene with minor negative impact on efficiency. Under optimal conditions Amds was proven to be faithful towards the production of amorpha-4,11-diene from FPP, converting ~ 80% of the substrate into amorpha-4,11-diene, ~ 5% amorpha-4,7 (11)-diene and ~3.5 % amorpha-4-en-7-ol together with 13 other sesquiterpenes in minute amounts.

Bertea et al.(2005) postulated that the main route to artemisinin is the conversion of amorpha-4,11-diene to artemisinic alcohol, which is further oxidized to artemisinic aldehyde (figure 3). The C11-C13 double bond in artemisinic aldehyde was then proposed to be reduced, giving dihydroartemisinic aldehyde which would upon

Chapter 2

34

further oxidation give dihydroartemisinic acid. The authors supported their conclusion by demonstrating the existence of amorpha-4,11-diene, artemisinic alcohol, artemisinic aldehyde and artemisinic acid together with the reduced forms of the artemisinin intermediates in leaf- and glandular trichome microsomal pellets, by direct extraction from leaves and through enzyme assays. Interestingly, they could not show any significant conversion of artemisinic acid into dihydroartemisinic acid regardless of the presence of cofactors NADH and NADPH thus strengthening the hypothesis that reduction of the C11-C13 double bond occurs at the aldehyde level. In view of these results it is very likely that artemisinic acid is a dead end product which can not be converted into artemisinin in contrast with some literature (Woerdenbag et al., 1990), unless reduced to dihydroartemisinic acid.

Recently, two research groups cloned the gene responsible for oxidizing amorpha-4,11-diene in three steps to artemisinic acid (Ro et al., 2006; Teoh et al., 2006) (figure 3). This enzyme, a cytochrome P450 named Cyp71av1, was expressed in Saccharomyces cerevisiae(S. cerevisiae) and associated to the endoplasmatic reticulum. The isolation and application of this cytochrome P450 is described further in the section Heterologous production in Saccharomyces cerevisiae. Further research that will clarify whether additional cytochrome P450s or other oxidizing enzymes are precent in the native biosynthetic pathway and where the reduction of the C11-C13 double bond occurs are still open fields of exploration.

Several terpenoids including artemisinin and some of its precursors and degradation products, has been found in seeds of A. annua (Brown et al., 2003). In its vegetative state, secretory glandular trichomes (Mueller et al., 2004) are the site of production of artemisinin. In a recent paper by Lommen et al. (2006) the authors show that the production of artemisinin is a combination of enzymatic and non-enzymatic steps. The authors followed the production of artemisinin and its precursors on a level per leaf basis.


35

The results showed that artemisinin is always precent during the entire life cycle of a leaf, from appearance to senescence and that the quantity is steadily increasing as would be expected for an end product in a biosynthetic pathway. Interestingly, the immediate precursor to artemisinin, dihydroartemisinic acid (Sy, 2002) was more abundant than other precursors, indicating that the conversion of dihydroartemisinic acid into artemisinin is a limiting step. It was also shown that dihydroartemisinic acid is not converted to artemisinin directly. The authors argue, in line with other literature (Sy, 2002), that this might be due to a temporary accumulation of the putative intermediate dihydroartemisinic acid hydroperoxide (figure 3). The observation that artemisinin levels continued to increase at the same time as the numbers of glandular trichomes decreased further supports the idea that the final step of artemisinin formation is non-enzymatic. Wallaart et al. (1999) were able to show that conversion of dihydroartemisinic acid to artemisinin is possible when using mineral oil as reaction solvent instead of glandular oil (figure 3). By adding dihydroartemisinic acid and chlorophyll a to mineral oil and exposing the mixture to air and light, a conversion of 12% after 120 hours was achieved. In absence of mineral oil a conversion of 26.8% could be achieved. Wallaart et al. (1999) were later able to show that the hypothesized intermediate between dihydroartemisinic acid and artemisinin, dihydroartemisinic acid hydroperoxide, could be isolated from A. annua and upon exposure to air for 24 hours at room temperature yielded artemisinin and dihydro-epi-deoxyarteannuin B (figure 3).

Genetic versus environmental regulation of artemisinin production

The genetic regulation of the biosynthesis of artemisinin is poorly understood on the single pathway level. The situation further complicates as there are several FPP synthase (Fpps) and 3-hydroxy-3-methylglutaryl CoA reductase (Hmgr) isoforms making optimization options more versatile and complex. The active drug component in A. annua was isolated in the 1970s but it is only during

Chapter 2

36

the last eight years that key enzymes in the committed biosynthetic pathway of artemisinin have been cloned and characterized (Chang et al., 2000; Mercke et al., 2000; Ro et al., 2006; Teoh et al., 2006; Wallaart et al., 2001) (figure 3). However, the genetic variation contributing to the level of artemisinin production has been investigated to some extent. The genetic variation is reflected in the existence of at least two chemotypes of A. annua. Wallaart et al. (2000) showed that plant specimens from different geographical origins had a different chemical composition of the essential oil during the vegetative period. The authors could distinguish one chemotype having a high content of dihydroartemisinic acid and artemisinin accompanied by a low level of artemisinic acid and a second chemotype represented by low artemisinin and dihydroartemisinic acid content together with a high level of artemisinic acid. With the aim to increase the artemisinin production the authors induced tetraploid specimens from normal high producing diploids using colchicine (Wallaart et al., 1999). This led to higher artemisinin content in the essential oil but to a 25% decrease in artemisinin yield per m2 leaf biomass.

Only a few studies have been performed investigating the effect of singular genes on artemisinin production. Wang et al. (2004) overexpressed a flowering promoting factor (fpf1) from Arabidopsis thaliana in A. annua and observed 20 days earlier flowering compared with the control plants but could not detect any significant change in artemisinin production. From this it can be concluded that the event of flowering has no effect on artemisinin biosynthesis, an idea supported by a later study performed by the authors in which the early flowering gene constans from A. thaliana was overexpressed in A. annua (Wang et al., 2006). In contrast, when an isopentenyl transferase gene from Agrobacterium tumefaciens (ipt) was overexpressed in A. annua, the content of cytokinins, chlorophyll and artemisinin increased 2-3 fold, 20-60% and 30-70% respectively (Sa et al., 2001). By overexpression of endogenous FPP in A. annua,Han et al. (2006) established a maximum of 34.4% increase in


37

artemisinin content corresponding to 0.9% of the dry weight. Similarly, a 2-3 fold increase in artemisinin production was obtained using a FPP from Gossypium arboreum (Chen et al., 2000).

To assess the genetic versus environmental contributions to artemisinin production, quantitative genetics was applied by Dealbays et al. (2001). Variance manifested in a phenotype or a trait such as a chemotype is the sum of the genetic and environmental variance. The genetic variance can in its turn be divided into additive genetic variance, dominance variance and epistatic variance. Additive variance is a representation of the number of different alleles of a trait, dominance variance the relation between dominant and recessive alleles and epistatic variance the relation between allele and alleles at different loci. Broad-sense heritability of a trait is defined as the variation attributed to genetic variance divided by the total variance in trait. Ferreira et al. (1995) estimated in their experiments a broad-sense heritability of up to 0.98. Delabays et al. (2001) confirm the broad-sense heritability of artemisinin to be between 0.95 and 1 and that the dominance variance of 0.31 was precent in the experiment. This implies that there are great variations between the same alleles which support, besides a genetic based existence of chemotypes, a mass-breading selection program of A.annua to produce artemisinin high yielding crop. With the breeding program CPQBA-UNICAMP aiming at improvement of biomass yields, rates between leaves and stem, artemisinin content, and essential oil composition and yield in A. annua, genotypes producing 1.69 to 2.01 g/m2 has been obtained (De Magalhaes et al., 2004).

Cell culture

One biotechnological research focus is to utilize hairy root cultures as a model of study and for production of artemisinin. Hairy roots are genetically and biochemically stable, are capable of producing a wide range of secondary metabolites, grow rapidly in comparison with the whole plant and can reach high densities (Flores et al., 1999;

Chapter 2

38

Shanks et al., 1999). It is an interesting approach but is currently hampered by the difficulties in scaling up the production to industrial proportions. Scaling-up of A. annua hairy root cultures has been shown to produce complex patterns of terpenoid gene expression pointing towards the difficulty of obtaining a homogeneously producing culture (Souret et al., 2003). Souret et al. compare in their study the expression levels of four key terpenoid biosynthetic genes, (figure 3) Hmgr, 1-deoxy-D-xylulose 5-phosphate synthase (Dxs), 1-deoxy-D-xyluose 5-phosphate reductoisomerase (Dxr) and Fpps, in three different culture conditions: shake flask, mist bioreactor and bubble column bioreactor. In shake flask conditions all key genes were temporally expressed but only Fpps had a correlation with artemisinin production. This is not surprising as the terpenoid cyclase has often proven to be the rate limiting step in a terpenoid biosynthetic pathway. Expressions of the genes in both bioreactor types were similar or greater than the levels in shake flask cultures. In the bioreactors, the transcriptional regulation of all the four key genes were affected by the position of the roots in the reactors, but there was no correlation with the relative oxygen levels, light or root packing densities in the sample zones. Medium composition and preparation has been proven to affect the production of artemisinin in hairy root cultures. Wen et al. (2002) showed that the ratio of differently fixed nitrogen in Murashige and Skoog medium (MS medium) had a great impact on artemisinin level. The optimal initial growth condition of 20 mM nitrogen in the ratio 5:1 NO3-/NH4+

(w/w) produced a 57% increase in artemisinin production compared to the control in standard MS medium. Weathers et al. (1997) determined optimal growth at 15 mM nitrate, 1.0 mM phosphate and 5% w/v sucrose with an 8 day old inoculum but the production of artemisinic acid was not detected using phosphate at higher concentrations than 0.5 mM. This implies that it is very difficult if even possible to optimize hairy root growth and terpenoid production at the same time, there has to be a trade off between biomass and product formation. As artemisinin is a secondary metabolite it is reasonable to assume that this compound will only be produced in


39

significant amounts when the primary needs of the tissue have been covered. An extended phase of biomass formation would mean a procrastinated production of secondary metabolites. Interestingly, Weathers et al. (1997) found that artemisinic acid was not detected when arteannuin B was produced (figure 3). They suggest that artemisinic acid is degraded by a peroxidase to arteannuin B which can be converted into artemisinin (Dhingra et al., 2001). This together with another observation that an oligosaccharide elicitor from the mycelial wall of an endophytic Colletotrichum sp. B501 promoted artemisinin production in A. annua hairy roots together with greatly increased peroxidase activity and cell death makes it tempting to see a peroxidase in the biosynthetic pathway from (dihydro)artemisinic acid to artemisinin (Wang et al., 2002). Dhingra et al. (2001) purified and characterized an enzyme capable of performing the peroxidation reaction converting arteannuin B to artemisinin (figure 3). Conversion was estimated to be 58% of the substrate on molar basis. Sangwan et al. (1993) was able to show conversion of artemisinic acid into arteannuin B and artemisinin using horse radish peroxidase and hydrogen peroxide on cell free extracts from immature A. annua leaves. On the other hand, it has been shown in chickory that the lactone ring formation in (+)-costunolide is dependent on a cytochrome P450 hydroxylase using germacrene acid as substrate (de Kraker et al., 2002).

Sugars are not only energy sources but also function as signals in plants (Loreti et al., 2006). Wethers et al. (2004) performed a study in which autoclaved versus filter sterilized media were used with the conclusions that filter sterilized media give higher biomass and more consistent growth results as well as better replicatable terpenoid production results although the yields of these secondary metabolites decreased. The authors explain the inconsistent results accompanying autoclaved media with variable hydrolysis of sucrose. By carefully choosing nutrient composition, light quality and bioreactor type the artemisinin production level can reach up to approximately 500 mg/l (Liu, et al., 1998; Liu et al., 1997). Ploidity

Chapter 2

40

is another factor to consider. De Jesus-Gonzales and Weathers produced tetraploid A. annua hairy roots by treating normal diploid parents with colchicine and thereby obtained a tetraploid hairy root producing six times more artemisinin than the diploid versions. Tetraploid plants have also been made using colchicine which led to a 39% increase in artemisinin production averaged over the whole vegetation period compared to diploid wild type plants (Wallaart, et al., 1999).

Heterologous biosynthesis

There are currently two main research strategies for production of artemisinin that are intensively pursued. One is to increase production in the plant by bioengineering or through breeding programs; the second strategy is to utilize microorganisms in artificial biosynthesis of artemisinin. The group focusing on plant improvement brings forward the advantages of low production costs and easy handling disregarding infestation and pest problems and additional costs for containment to prevent ecological pollution. The group favoring heterologous production of artemisinin in microorganisms admits higher production costs at the moment compared with artemisinin isolated from the plant but points to the advantages of efficient space versus production ratio, complete production and quality control and continuous supply of artemisinin possible only with sources not dependent on uncontrollable factors such as weather. In the two following chapters we give examples of the progress of the heterologous production of isoprenoid and artemisinin precursors in microorganisms.

Heterologous production in Escherichia coli

Of the two isoprenoid biosynthetic pathways that exist (figure 3), the Dxp is used by most prokaryotes for production of IPP and dimethylallyl diphosphate (DMAPP) (Boucher et al., 2000; Rohdich, et al., 2002). With the knowledge of the genes involved in the DXP


41

pathway available, several groups have studied the impact of changed expression levels of these genes on the production of reporter terpenoids. Farmer et al. reconstructed the isoprene biosynthetic pathway in Escherichia coli (E. coli) to produce lycopene which was used as an indication of increase or decrease in isoprenoid production levels (Farmer et al., 2001). By overexpressing or inactivating enzymes involved in keeping the balance of pyruvate and glyceraldehyde 3-phosphate (G3P) the authors established that directing flux from pyruvate to G3P increased lycopene production making the available pool of G3P the limiting precursor to isoprenoid biosynthesis. Kajiwara et al. (1997) showed that overexpression of IPP lead to increased production of their terpenoid reporter molecule beta carotene. Kim et al. (2001) investigated the influence of Dxs, Dxr, plasmid copy number, promoter strength and strain on production of the reporter terpenoid lycopene and were able to show a synergistic positive effect upon overexpression of both genes. These kinds of strategies have all led to a moderate increase in production of terpenoid reporter molecules. Martin et al. (2003) hypothesized that the limited increase in isoprene production could be attributed to unknown endogenous control mechanisms. By introducing the heterologous mevalonate pathway from S. cerevisiae into E. coli these internal controls were bypassed and isoprenoid precursors reached a toxic level. Introduction of a codon optimized Amds alleviated this toxicity and led to production of amorpha-4,11-diene at the level of 24 μg caryophyllene equivalent/ml (Martin et al., 2003; Newman et al., 2006). Genes, such as transcriptional regulators, that are not directly involved in the isoprenoid biosynthetic pathway have also been shown to have a similar impact on production levels of terpenoid reporter molecules (Kang et al., 2005). This can be expected as the isoprenoid biosynthetic pathway is tightly intertwined with the energy metabolism of the cell. The design strategy of the construct used can have a great influence on precursor production as showed by Pfleger et al. (2006). By tuning intergenic regions in the mevalonate operon constructed by Martin et al. (2003), a seven fold

Chapter 2

42

increase in mevalonate production was recorded compared with the starting operon conditions. Brodelius et al. (2002) went a step beyond manipulating isolated genes, singular or multiple, in the biosynthesis of isoprenoids. By fusing Fpps isolated from A. annuaand epi-aristolochene synthase from tobacco, the extreme proximity and therefore very short diffusion path led to a 2.5 increase in epi-aristolochene compared to solitary epi-aristolochene synthase. Heterologous production of cyclized terpenoids is efficient but the following modification to form oxygenated plant terpenoids in E.coli seems to be a great bottleneck. Carter et al. (2003) engineered GPP biosynthesis coupled with the monoterpene cyclase limonene synthase, cytochrome P450 limonene hydroxylase, cytochrome P450 reductase and carveol dehydrogenase in E. coli with the expectation of producing the oxygenated limonene skeleton (-)-carvone. Production of the unoxygenated intermediate limonene reached 5 mg/l but no oxygenated product was detected. The authors argue that this limitation could be due to cofactor limitations and membrane structural limitations in E. coli compared to plants. Hence, several research groups have turned to yeast for heterologous expression of complex biosynthetic pathways.

Heterologous production in Saccharomyces cerevisiae

Fungi use the mevalonate pathway to produce all their isoprenoids. Lessons learned on manipulation of the genes involved in the yeast mevalonate pathway, were useful to increase the production of isoprenoids in E. coli as discussed above. Jackson et al. (2003) used epi-cedrol synthase converting FPP as a reporter gene for ispoprenoid production. By overexpressing a truncated version of Hmgr in a S. cerevisiae mutant (Upc2-1, upregulates global transcription activity) taking up sterol, which is a major terpenoid product, from the media an increase from 90μg/l to 370μg/l epi-cedrol was obtained. Overexpression of a native Fpps gene did not, however, improve levels of epi-cedrol. As in the attempt of heterologous production of the oxygenated terpenoid epi-cedrol


43

(Carter et al., 2003) in E. coli, an attempt to reconstruct early steps of taxane diterpenoid (taxoid) metabolism in S. cerevisiae produced taxadiene but did not proceed with cytochrome P450 hydroxylation steps (Dejong et al., 2006). Membrane structural limitations or co-factor limitations such as NADPH can not explain this result. The authors discussed that poor expression of the heterologous plant cytochrome P450 genes might be an explanation to this pathway restriction. Another angle mentioned by the authors is a possible inefficient coupling and interaction between the endogenous yeast NADPH-cytochrome P450 reductase and the plant cytochrome P450 hydroxylase, thus severely limiting the transfer of electrons to the cytochrome P450 hydroxylase and as a consequence premature termination of the pathway. Ro et al. (2006) introduced several genetic modifications in S. cerevisiae and were able to produce the oxygenated terpenoid artemisinic acid at 100 mg/l titre. This was achieved by optimized oxygen availability, downregulation of squalene synthase (Erg9) thus reducing endogenous consumption of the FPP pool, introduction of the Upc2-1 mutation, overexpression of Fpps and a catalytic form of Hmgr, inducible expression of Amds, cytochrome P450 71av1 and a cytochrome P450 reductase from A.annua. More than 50 mg/l amorphadiene was produced in yeast engineered for overexpression of truncated Hmgr and Amds in a Upc2-1 yeast mutant genetic background. An additional two to threefold increase in amorphadiene level was obtained through knock out of squalene synthase but a marginal increase was harvested with additional overexpression of Fpps. Teoh et al. (2006) showed that oxygenation of amorphadiene to artemisinic alcohol and artemisinic alcohol to artemisinic acid was possible at proof of principle levels using cytochrome P450 71av1 and a cytochrome P450 reductase from Arabidiopsis thaliana (Urban et al., 1997). Takahashi et al. (2006) chose a similar strategy to create a yeast platform for production and oxygenation of terpenes. S. cerevisiaemutated in squalene synthase (Erg9) and capable of efficient aerobic uptake of ergosterol (Sue) from the culture media, produced 90mg/l farnesol, which is the dephosphorylated form of FPP and is

Chapter 2

44

unaccessible for cyclization through terpene synthases. This mutant was, when engineered with various single terpene synthases, capable of producing around ~80- ~100 mg/l sesquiterpene varying with terpene synthase introduced. After additional engineering with hydroxylases, up to 50 mg/l hydroxylated terpene and 50 mg/l unmodified terpene product were obtained. Knocking out a phosphatase (Dpp1) known to dephosphorylate FPP (Faulkner et al., 1999) and additional upregulation of the catalytic activity of Hmgr did not yield an increase in terpene production compared to the Erg9/Sue yeast mutant. The authors note that a larger part of the farnesol is phosphorylated in a Dpp1 mutant and as FPP function as a negative feedback signal on the mevalonate pathway it suppress the flux of carbon through the isoprene pathway (Gardner et al., 1999; Takahashi et al., 2006). Inserting a terpene cyclase diverts the pool of FPP and relieves the feedback inhibition which leads to an increase in carbon flux through the pathway almost matching the Erg9/Sue yeast mutant. In the Erg9/Sue mutant, a low but continuous flow through the mevalonate pathway leads to higher production of terpenoids. Takahashi et al. (2006) also illustrate the importance of design strategy of the expression vectors for optimal terpene production. Physically separating the cytochrome P450 reductase and the cytochrome P450 hydroxylase, led to a very low yield of oxygenated terpenoid. On the other hand, expression vectors where the reductase preceded the hydroxylase gene on the same plasmid yielded approximately 50% coupling of oxygenation to hydrocarbon. Physically linking the terpene synthase with the hydroxylase was unsuccessful using both N-terminal and C-terminal fusion. Lindahl et al. (2006) showed that there are great differences in the production of amorphadiene depending on genomic or episomal expression. The authors compared production of amorpha-4,11-diene using the terpenoid synthase cloned in the high-copy number galactose inducible yeast plasmid pYeDP60 with the terpenoid synthase using the same galactose inducible promoter integrated into the genome of S. cerevisiae CEN PK113-5D. It was found that the yeast with an integrated Amds grew at the same rate as the wild type while the


45

yeast carrying Amds episomaly had a slightly lower growth rate, yet the episomal system produced 600μg/l amorpha-4,11-diene compared to 100 μg/l for the integrated system. This is an expected result which shows that in the case of integrated Amds the enzyme activity is the limiting factor while in the episomal system substrate availability is the limiting factor.

Growth of Artemisia annua in the fields and in controlled environments

Studies have been made where the intrinsic capacity of A. annua to produce artemisinin under various environmental conditions were explored. Ram et al. (1997) performed a study in which A. annuawas grown with varying plant densities during the winter-summer season of one year in semiarid-subtropical climate with no interculture and no fertilization. At a population density of 2.22x105

plants/ha 7.4 kg of artemisinin were obtained and 91 kg of essential oil. By increasing the plant density 2, 4 and 8-fold an increase of artemisinin by 1.5, 2 and 2.5 fold was observed at the same oil yield level. Interestingly, the suppression of weeds was positively correlated with the increase in artemisinin production. Weed, however, does not seem to be a trigger for artemisinin production as treatment of A. annua with herbicides did remove weeds but did not influence artemisinin yields (Bryson et al., 1991). Kumar et al. (2004) showed that multiple harvesting of A. annua grown in the subtropical Indo-Gangatic plains, not surprisingly, increased the total yield of artemisinin but also increased the production of artemisinin in leaves as averaged over the separate sampling events. This trend was more expressed the later in the year the seeds were sown, confirming a study performed by Ram et al. (1997). The effect of post-harvest treatment of A. annua on artemisinin content was investigated by Laughlin (2002) in a study using A. annua grown and harvested in temperate maritime environment in Tasmania. The experiments included drying of the cut off plants in situ, in the shadow, indoors in the dark or in a 35°C oven (used as comparison

Chapter 2

46

base). Drying in situ did not give any concentration difference in artemisinin content compared to oven treatment. The authors noted a trend for sun-, shade- and dark drying for 21 days to give higher artemisinin levels than oven drying although artemisinic acid levels were unaffected.

Under greenhouse controlled conditions, Ferreira (2007) investigated the impact of acidity and macronutrient deficiency on biomass and artemisinin yield. Acidic soil and low levels of nitrogen, phosphor and potassium reduced, as expected, leaf biomass to 6.18 g/plant. Providing lime to increase pH and addition of the macronutrients nitrogen, phosphor and potassium gave a biomass of 70.3 g/plant. Potassium deficiency was shown to have the least negative effect on biomass accumulation and most positive effect on artemisinin production. Plants grown under potassium deficient conditions were compared with plants grown under full addition of lime and macronutrients. This comparison did not detect any significant change in artemisinin production between the two growth conditions. The author concludes that under mild potassium deficiency conditions, similar production of artemisinin can be obtained per ha as when fertilizing the soil with potassium. Potassium fertilization can thus be omitted in acidic soil growth conditions, decreasing the production costs as stated by the author but would also decrease the environmental pressure.

Synthesis of artemisinin, derivatives and new antiplasmodial drugs

Ever since artemisinin was isolated as the active compound against malaria, organic chemists have been trying and succeeding to produce the drug in the reaction flask. This has been performed with variable success but the general conclusion is still that it is a great scientific achievement but economically not attractive. A recent synthetic route to artemisinin involves 10 reaction steps from (+)-isolimonene to (+)-artemisinin with a final yield of a few percent


47

(Yadav et al., 2003). Yet this result is considered a success in terms of yield and stereochemistry precision. In contrast, conversion of artemisinic acid into artemisinin is simple and can be done with photooxygenation in organic solvent (Roth et al., 1989). Sy et al. (2002) describe in their study the role of the 12-carboxylic acid group in spontaneous autooxidation of dihydroartemisinic acid to artemisinin. The mechanism is further developed in the accompanying paper by the authors (Sy et al., 2002). Artemisinin, however, has very poor solubility in both oil and water and is despite its antiplasmodial activity therefore not suitable as a drug. The development of artemisinin derivatives and completely synthetic analogues is described in a review by Ploypradith (2004). Among the first tries to improve the solubility of artemisinin were to replace the ketone with other bigger polar groups forming ester derivates of artemisinin. Depending on the attached groups, the first generation derivates showed solubility in either oil or water. The derivates sodium artesunate and artelinic acid are still in use due to their efficiency in clearing severe malaria infections. However, these first generation derivates are labile in acid environment, have a short half life and some derivates has shown to have neurotoxic effects. The second generation of semi-synthetic analogues were produced from artemisinin or artemisinic acid with the goals of improvement in metabolic and chemical stability, bioavailability and half-life. Two main streams were developed in the second generation of semi-synthetics. One group retained the acetal C10-oxygen, the other line developed further on the line converting this oxygen to a carbon to increase the stability in acidic conditions. Of these two groups there are monomers and dimers. The dimers are interesting not only because they have a high antiplasmodial activity, but also because of their antineoplastic features.

Artemisinin with its crucial endoperoxide bridge is not the only natural compound exhibiting antiplasmodial activity. An example of the biosynthesis of antiplasmodial endoperoxidic compounds is plakortin, a simple 1,2-dioxane derivative, which is produced by the

Chapter 2

48

marine sponge Plakortis simplex (Fattorusso et al., 2006). This compound shows activisty against chloroquine-resistant strains of Plasmodium falciparum (P. falciparum) at submicromolar level.

Several synthetic simplifications have been made as the knowledge of the mode of action of artemisinin has developed. In a review by Ploypradith (2004), selected strategies are reported. One line is to omit the lactone ring which is considered to be less important if at all for antiplasmodial activity. Molecules that completely abandon the structure of artemisinin and its precursor only retaining the peroxide bond as a crucial functional pharmacophore are numerous. These molecules are easy to make but unfortunately display significantly reduced activity against malaria compared with artemisinin. As discussed in the introduction they have a short half-life and poor chemical stability. A further dimension added in the synthesis of synthetic aniplasmodials was the idea to add multiple endoperoxide bridges within a molecule ring rather than adding them up as dimers with a linker in between. These tetraoxacykloalkanes showed an several fold increase in efficiency against malaria compared to artemisinin yet had a lower toxicity in mouse models. Design and synthesis of selected tetraoxanes are described in an article by Amewu et al. (2006).

Analytics

The detection and structural elucidation of terpenes has been hampered by the often very low amounts and complex mixtures formed in plants. The honing of extraction methods and analytical methods has increased the ease and speed with which these problems can be solved. The choice of extraction protocol greatly influence the yield and composition of the isolated product as well as the cost and time factors (Christen et al., 2001). Peres et al. (2006) compare soxhlet, ultrasound-assisted and pressurized liquid extraction of terpenes, fatty acids and vitamin E from Piper gaudichaudianum Kunth. The authors conclude that the method pressurized liquid


49

extraction decrease the total time of extraction, solvent use and handling compared to the other two methods. Furthermore, it was determined that pressurized liquid extraction was more efficiently extracting terpenes than the other two methods. Lapkin et al. (2006) compare extraction of artemisinin using hexane, supercritical carbon dioxide, hydrofluorocarbon HFC-134a, several ionic liquids and ethanol. Hexane was found to be simple and at a first glance most cost efficient but is characterized by lower rates and efficiency compared to all other methods including safety and environmental impact issues. The new techniques based on supercritical carbon dioxide, hydrofluorocarbon HFC-134a and ionic liquids consistently showed faster extraction cycles with higher recovery in addition to enhanced safety and decreased negative impact on the environment compared to hexane and ethanol extraction. With some process optimization, the authors predict that ionic liquid and HFC-134a extraction can compete with hexane extraction also on economical terms. Christen et al. (2001) compare in their review article the extraction techniques supercritical fluid extraction, pressurized solvent extraction and microwave-assisted extraction and the detection methods gas chromatography, tandem mass spectrometry, HPLC-UV, -EC and -MS as well as ELISA and capillary electrophoresis. The use of evaporative light scattering detector is mentioned as a tool for detection of non-volatile non-chromophoric compounds. Common to all these methods is the trend toward mild operating conditions to avoid degradation of the analytes, isolation of one compound in complex mixtures and time and price reduction compared to traditional extraction methods. ELISA is accurate and is usable for screening of large plant populations but is laboursome and expensive compared to standard GC and HPLC based methods (Ferreira et al., 1996). It is likely that this method will win stronger support in assessing the drug susceptibility of P. falciparum (Kaddouri et al., 2006). A simple, fast and selective method of quantification of artemisinin and related compounds was developed by Van Nieuwerburgh et al. (2006). This method makes use of HPLC-ESI-TOF-MS/MS technology and has a recovery of >97% for

Chapter 2

50

all measured analytes. Peng et al. (2006) compared the use of GC-FID and HLPC-ELSD for detection of artemisinin in leaves. Both methods are valuable for routine measurements as they are cheap, easy to use and do not require derivatization of artemisinin for detection. Both methods had a high sensitivity at ng level and produced reproducible results of artemisinin from field plants with a correlation coefficient of r2 = 0.86 between the two methods. Another interesting simple and rapid method circumventing the problems with thermolability, lack of chromophoric or fluorophoric groups, low concentration in vivo and interfering compounds inplanta of artemisinin detection is the method developed by Chen et al. (2002). Artemisinin is converted on-line to the strongly absorbing compound Q292 through treatment with NaOH. The obtained product is analyzed with capillary electrophoresis in 12 minutes, allowing a sampling frequency of 8 h-1. With this work, Chen et al. (2002) show that it is possible to determine artemisinin content based on the unstable UV-absorbing compound Q292 thus omitting the traditional time-consuming step of acidic conversion of Q292 to the stable UV-absorbing compound Q260 before analysis. A HPLC-MS method in selective ion mode developed by Wang et al. (2005) is another interesting cheap, sensitive and fast method for detection and quantification of artemisinin in crude plant extracts. The obtained linearity of detection in this method is about 5-80 ng/ml for artemisinin with an analyze time of 11 min per sample.

An old method revived is the use of thin layer chromatography plates for detection of sesquiterpenoids (Bhandari et al., 2005; Klayman et al., 1984). While this kind of detection is qualitative and preferably used as quick determination of yes/no cases, more comprehensive and qualitative methods are needed for research purposes.

Ma et al. (2007) made a finger print of the volatile oil composition of A. annua by using two dimensional gas chromatography time-of-flight mass spectrometry. With this method, approximately 700 unique peaks were detected thereof 303 tentatively identified (Ma, et


51

al., 2007). As a comparison, only 61 peaks could be detected using GC. This type of comprehensive metabolic fingerprinting will ease detection of genes directly or indirectly relevant for the biosynthesis of artemisinin in experiments utilizing gene upregulation or downregulation mechanisms.

There is some discussion about synergistic effects on clearing of the parasite P. falciparum from infected patients using extracts from A.annua. Bilia et al. (2002) describe the importance of flavonoids in interaction between artemisinin and hemin. Hemin is thought to play a role in activation of artemisinin. It is thus valuable to develop a method which can analyze artemisinin and flavonoids simultaneously. Bilia et al. (2006) developed a method based on HPLC/diode-array-detector/MS delivering just that.

Medicinal use

The mode of action of artemisinin is subject to intense research (Drew et al., 2006; Hoppe et al., 2004; Krishna et al., 2004; Messori et al., 2004; O'Neill et al., 2005; Posner et al., 2004; Rafiee et al., 2005; Schmuck et al., 2002). Currently, the hypothesis supporting radical ion formation from artemisinin on the peroxide bridge is favored.

Traditionally, artemisinin is administered as a tea infusion. With the advent of combination therapies using artemisinin as an isolated compound it is necessary to compare the kinetic characteristics of each delivery method. Rath et al. (2004) studied the pharmacokinetics and bioavailability of artemisinin from tea and oral solid dosage forms. Interestingly, artemisinin was absorbed faster from herbal tea preparations than from oral solid forms, supporting the importance of flavonoids as synergistic factors. Nevertheless, bioavailability was similar in both treatments. As only about 90 mg artemisinin was contained in 9 g A. annua and as uptake of artemisinin through the human gut is very poor, only about 240

Chapter 2

52

ng/ml was detected in plasma, a tea infusion is not recommended by the authors as a replacement for modern formulations in malaria therapy. This confirms the study of pharmacokinetics of artemisinin performed by Duc et al. (1994). Duc et al. (1994) propose to increase the dose of artemisinin until adequate plasma levels are reached to compensate for poor bioavailability and rapid elimination as no adverse effects were detected. This might prove a risky strategy as artemisinin-induced toxic brainstem encephalopathy has been observed in a patient treated for breast cancer with artemisinin (Panossian et al., 2005). The adverse effects were reversible and no permanent damage could be detected. Toxicity of antimalarials including artemisinin derivatives is described in a review article by Taylor et al. (2004). Mueller et al. (2004) studied the efficacy and safety of the use of A. annua as tea against uncomplicated malaria in a pilot study. Treatments were efficient but still less efficient compared to the traditional quinine, an average of 74% were cleared after 7 days of treatment compared to 91% treated with quinine. As the authors note, recrudescence rates were high in the groups treated with artemisinin and they therefore recommend combination therapies which is in line with the recommendation from WHO. However, the choice of combination partner in the combination therapies is a delicate question which is exemplified in the study of Sisowath et al. (2005). In a recent review article the mechanism behind antimalarial drug resistance is covered (White, 2004). Interestingly, resistance can be reversed (Henry et al., 2006). It is obvious that the clearance of the parasite through tea preparations will depend on the amount of artemisinin present in the plant as only approximately 40% of the available artemisinin in the plant was recovered in tea infusions as shown in another study by Mueller et al. (2000). Here it was demonstrated that malaria infested patients who were given tea preparations for 2-4 days showed a recovery of 92% within 4 days, a remarkable improvement compared with the previous mentioned study (Mueller et al., 2004).


53

An overview of older (up to 1999) artemisinin derivatives is given in the article by Dhingra et al. (2000). All these derivatives were developed with the aim of obtaining a more efficient remedy against malaria. However, more recently artemisinin and derivatives have been attributed other intriguing functions than antiplasmodial activities. Romero et al. (2006) describe in a study on flaviviruses the antiviral property of artemisinin. Zhou et al. (2005) observed the derivate 3-(12- -artemisininoxy) phenoxyl succinic acid (SM735) to be strongly immunosuppressive in vitro and in vivo. Artemisinin derivatives have also been shown to have strong antineoplastic properties (Disbrow et al., 2005; Efferth, 2006; Efferth et al., 2002; Liu et al., 2005).

Pharmacokinetics

A characteristic of artemisinin and its related endoperoxide drugs is the rapid clearance of parasites in the blood in almost 48 hours. Titulaer et al. (1990) obtained pharmacokinetic data for the oral, intramuscular and rectal administration of artemisinin to volunteers. Rapid but incomplete absorption of artemisinin given orally occurs in humans with a mean absorption time of 0.78 h with an absolute bioavailability of 15 % and relative bioavailability of 82%. Peak plasma concentrations at a given dose are reached after 1-2 h and the drug is eliminated after 1 to 3 hours. The mean residence time after intramuscular administration was three times that when given orally. Other routes of administration, for example rectal or transdermal, are of limited success, but for the treatment of convulsive malaria in children artemether in a rectal formulation is favored. Artesunate acts as a prodrug that is converted to dihydroartemisinin. When given orally the first pass mechanism in the gut wall takes places metabolizing half of the administered dose. Oral artemether is rapidly absorbed reaching maximum blood levels (Cmax) within 2-3 hours. Intramuscular artemether is rapidly absorbed reaching Cmaxwithin 4-9 hours. It is metabolized in the liver to the demethylated derivative dihydroartemisinin. The elimination is rapid, with a half-

Chapter 2

54

life time (T1/2) of 4 hours. In comparison, dihydroartemisinin has a T1/2 of more than 10 hours. The degree of binding to plasma proteins varies markedly according to the species considered. The binding of artemether to plasma protein was 58% in mice, 61% in monkeys and 77% in humans. Radioactive labelled artemether was found to be equally distributed in plasma as well as in red blood cells indicating an equal distribution of free drug between cells and plasma.

From the toxicological point of view artemisinin seems to be a safe drug for the use in humans. In animal tests neurotoxicity has been documented, but as yet this side effect has not been reported in humans (Merali et al., 1991). A major disadvantage of the artemisinin drugs is the occurrence of recrudescence when given in short monotherapy. So far no resistance has been observed clinically although it has been induced in rodent models in vivo. The mechanism of action is different from the other clinically used antimalarials. Artemisinin drugs act against the early trophozoite and ring stages, they are not active against gametocytes, and it affects blood- but not liver-stage parasites. The mode of action is explained by haem or Fe2+, from parasite digested haemoglobin, catalysing the opening of the endoperoxide ring and forming free radicals. Malaria parasites are known to be sensitive to radicals because of the lack of enzymatic cleaving mechanisms. The mechanism of action and the metabolism of reactive artemisinin metabolites is shown in figure 6.


55

O

O

CH3

O

O

O

CH3

CH3

O

O

CH3

O+

OFe

O

CH3

CH3

Haem (Fe2+)

O

O

CH3

FeOO

O

CH3

CH3

O

O

CH3

HOOFe

O

CH3

CH3

O

O

CH3

O

CH3

CH3

O Alkylation of proteins

Formation of metabolites

Figure 6. Mechanism of action of artemisinin drugs. Active metabolites and formation of reactive epoxide intermediates (van Agtmael et al., 1999).

Chapter 2

56

Drug delivery

Drug delivery of artemisinin and its derivatives is not that easy as known for intracellular microorganisms like Leishmania sp., Mycobacterium tuberculosis or Listeria monogynes. Plasmodium falciparum and related species are facultative intracellular parasites which mainly persist in erythrocyte as host cells. Drug targeting of infected erythrocytes is not well known and it seems not to be a major area of interest for pharmaceutical technology to identify new strategies to deliver artemisinin or other antiplasmodial drugs to this target site. Literature search revealed no publication using liposomes, microemulsions, nanoemsulsions, microparticles or nanoparticles for targeting or drug delivery. Most formulation strategies have been focused on the improvement of the poor solubility of artemisinin (< 5 mg/l H2O). One interesting approach has been published in detail documenting the approach to increase solubility with cyclodextrines. Cyclodextrines are cyclic oligosaccharides consisting of six, seven or eight glucose molecules forming -, -, or -cyclodextrine, respectively. Cyclodextrines form pores with an inner diameter ranging from 0.5 to 0.8 nm where lipophilic drugs may be incorporated, thereby increasing their distribution in water. While the lipophilic compound is shielded inside, hydroxyl groups on the outer surfaces create an overall hydrophilic character for this inclusion complex. For experimental purposes, artemisinin has been formulated with different cyclodextrines to improve its solubility and oral absorption leading to increased bioavailability (Wong et al., 2003). Solubility diagrams indicated that the complexation of artemisinin (85%, 40%, and 12%, -, -, or -cyclodextrine respectively) and the three different types of cyclodextrines occurred at a molar ratio of 1:1, and showed a remarkable increase in artemisinin solubility (Wong et al., 2003). In a bioavailability study from the same authors -, or -cyclodextrines seem to be superior to commercial Artemisinin 250 and increased oral bioavailability with a mean of 782 ng h/ml to 1329 and 1131 ng h/ml ( -, or -cyclodextrine respectively). But still the poor solubility was a critical


57

parameter for significantly improved oral bioavailability (Wong et al., 2001).

Conclusion

Artemisinin is a potent antimalarial drug belonging to the chemical class of sesquiterpenoid endoperoxide lactones. Its poor solubility in water and organic phases has led to a focus on development of derivatives towards increased solubility, metabolic and chemical stability and bioavailability (Ploypradith, 2004). The first generation of artemisinin derivatives had as a common feature the replacement of the ketone with bigger polar groups to form ester derivatives (figure 1). Among these, sodium artesunate and artelinic acid are still in use. Unfortunately other common features of the first generation artemisinin derivatives are instability in acid environment and short half life. Some derivatives also have a neurotoxic effect. The second generation of semi-synthetic artemisinin derivatives target improved metabolic and chemical stability, bioavailability and half-life. In parallel with the progressing understanding of the mode of action of artemisinin, synthetic simplified antimalarial compounds have been developed. Several promising candidates based on synthetic simplified molecules containing multiple peroxide bridges within one ring which show higher activity against malaria and lower toxicity compared with artemisinin have been reported (Ploypradith, 2004).

Two genes have been isolated from the biosynthetic pathway of artemisinin: the first committed step in the pathway, amorpha-4,11-diene synthase and the next enzyme in the pathway, the cytochrome P450 71av1 which catalyzes three consecutive oxygenation steps on the amorphane sesquiterpene skeleton yielding artemisinic acid (figure 3, figure 4) (Ro et al., 2006; Teoh et al., 2006). This opens up for molecular biotechnology strategies aiming towards artificial biology making use of heterologous gene expression in optimized hosts and improvement of artemisinin yield in transgene A. annua

Chapter 2

58

through genetic engineering. With the knowledge of nucleotide sequences, protein functions and characteristics, evolution of the genes identified in the biosynthetic pathway is a possible and logical next step to follow for increased levels of the artemisinin precursors amorpha-4,11-diene and oxygenated forms thereof. Great improvement in yield of amorpha-4,11-diene and other early precursors has been made with the aid of genetic engineering and optimization of culture conditions. There are currently two main research lines followed in parallel with the third line favoring artificial biology strategies with the aim of increased artemisinin production compared to the wild type plant: The use of cell cultures, a field which combines culture optimization and genetic engineering and the second line which employs traditional plant breeding through which the genetic dominance over environmental impact on artemisinin production is exploited. All strategies show potential for substantial improvement and it is currently not settled which if any approach is the better one in terms of economy, environmental impact, yield, safety and production flow. The recent developments in detection and separation technologies of terpenoids should aid a swift progress in screening mutants and complex networks in which the artemisinin biosynthesis pathway is embedded.

The traditional administration of artemisinin as a tea of the plant A. annua is a cheap, easily accessible source for malaria plagued countries but is an unreliable cure due to the fact that artemisinin level in planta is very low and varies considerably between plants and batches. Additionally, absorption through the human gut is rapid but inefficient and liver induction of cytochromes P450s will not allow repeated drug courses. The most efficient administration is intramuscular injection as the drug then has a mean residence time three times longer than orally administrated drug. As intramuscular administration requires medical personnel, is painful and generally disliked by patients, other delivery strategies is an urgent research field together with techniques increasing the solubility of artemisinin such as the use of cyclodextrines (Wong et al., 2003).


59

References

Amewu, R., Stachulski, A. V., Ward, S. A., Berry, N. G., Bray, P. G., Davies, J., et al. (2006). Design and synthesis of orally active dispiro 1,2,4,5-tetraoxanes; synthetic antimalarials with superior activity to artemisinin. Org Biomol Chem, 4(24), 4431-4436.

Benedict, C. R., Lu, J. L., Pettigrew, D. W., Liu, J., Stipanovic, R. D., & Williams, H. J. (2001). The cyclization of farnesyl diphosphate and nerolidyl diphosphate by a purified recombinant delta-cadinene synthase. Plant Physiol, 125(4), 1754-1765.


Bhandari, P., Gupta, A. P., Singh, B., & Kaul, V. K. (2005). Simultaneous densitometric determination of artemisinin, artemisinic acid and arteannuin-B in Artemisia annua using reversed-phase thin layer chromatography. Journal of separation science, 28(17), 2288-2292.

Bilia, A. R., Lazari, D., Messori, L., Taglioli, V., Temperini, C., & Vincieri, F. F. (2002). Simple and rapid physico-chemical methods to examine action of antimalarial drugs with hemin: its application to Artemisia annua constituents. Life Sci, 70(7), 769-778.

Bilia, A. R., Melillo de Malgalhaes, P., Bergonzi, M. C., & Vincieri, F. F. (2006). Simultaneous analysis of artemisinin and flavonoids of several extracts of Artemisia annua L. obtained from a commercial sample and a selected cultivar. Phytomedicine, 13(7), 487-493.

Chapter 2

60

Boucher, Y., & Doolittle, W. F. (2000). The role of lateral gene transfer in the evolution of isoprenoid biosynthesis pathways. Mol Microbiol, 37(4), 703-716.

Bouwmeester, H. J., Kodde, J., Verstappen, F. W., Altug, I. G., de Kraker, J. W., & Wallaart, T. E. (2002). Isolation and characterization of two germacrene A synthase cDNA clones from chicory. Plant Physiol, 129(1), 134-144.


Brodelius, M., Lundgren, A., Mercke, P., & Brodelius, P. E. (2002). Fusion of farnesyl diphosphate synthase and epi-aristolochene synthase, a sesquiterpene cyclase involved in capsidiol biosynthesis in Nicotiana tabacum. Eur J Biochem, 269(14), 3570-3577.

Brown, G. D., Liang, G. Y., & Sy, L. K. (2003). Terpenoids from the seeds of Artemisia annua. Phytochemistry, 64(1), 303-323.

Bryson, C. T., & Croom, E. M., Jr. (1991). Herbicide inputs for a new agronomic crop , annual wormwood ( Artemisia annua ). Weed Technology 5(1), 117-124.

Cane, D. E., & Watt, R. M. (2003). Expression and mechanistic analysis of a germacradienol synthase from Streptomyces coelicolor implicated in geosmin biosynthesis. Proc Natl Acad Sci U S A, 100(4), 1547-1551.

Carter, O. A., Peters, R. J., & Croteau, R. (2003). Monoterpene biosynthesis pathway construction in Escherichia coli. Phytochemistry, 64(2), 425-433.

Caruthers, J. M., Kang, I., Rynkiewicz, M. J., Cane, D. E., & Christianson, D. W. (2000). Crystal structure determination of aristolochene synthase from the blue cheese mold, Penicillium roqueforti. J Biol Chem, 275(33), 25533-25539.

Chang, Y. J., Song, S. H., Park, S. H., & Kim, S. U. (2000). Amorpha-4,11-diene synthase of Artemisia annua: cDNA isolation and bacterial expression of a terpene synthase


61

involved in artemisinin biosynthesis. Arch Biochem Biophys, 383(2), 178-184.

Chen, D.-H., Ye, H.-C., & Li, G.-F. (2000). Expression of a chimeric farnesyl diphosphate synthase gene in Artemisia annua L. transgenic plants via Agrobacterium tumefaciens-mediated transformation. Plant Science, 155(2), 179-185.

Chen, H. L., Wang, K. T., Pu, Q. S., Chen, X. G., & Hu, Z. D. (2002). On-line conversion and determination of artemisinin using a flow-injection capillary electrophoresis system. Electrophoresis, 23(17), 2865-2871.

Chen, X. Y., Wang, M., Chen, Y., Davisson, V. J., & Heinstein, P. (1996). Cloning and heterologous expression of a second (+)-delta-cadinene synthase from Gossypium arboreum. JNat Prod, 59(10), 944-951.

Christen, P., & Veuthey, J. L. (2001). New trends in extraction, identification and quantification of artemisinin and its derivatives. Curr Med Chem, 8(15), 1827-1839.

Colby, S. M., Alonso, W. R., Katahira, E. J., McGarvey, D. J., & Croteau, R. (1993). 4S-limonene synthase from the oil glands of spearmint (Mentha spicata). cDNA isolation, characterization, and bacterial expression of the catalytically active monoterpene cyclase. J Biol Chem, 268(31), 23016-23024.

de Kraker, J. W., Franssen, M. C., Joerink, M., de Groot, A., & Bouwmeester, H. J. (2002). Biosynthesis of costunolide, dihydrocostunolide, and leucodin. Demonstration of cytochrome p450-catalyzed formation of the lactone ring precent in sesquiterpene lactones of chicory. Plant Physiol, 129(1), 257-268.

De Magalhaes, P. M., Pereira, B., & Sartoratto, A. (2004). Yields of antimalarial Artemisia annua L. species. Acta Horticulturae 629, 421-424.

Dejong, J. M., Liu, Y., Bollon, A. P., Long, R. M., Jennewein, S., Williams, D., et al. (2006). Genetic engineering of taxol

Chapter 2

62

biosynthetic genes in Saccharomyces cerevisiae. Biotechnol Bioeng, 93(2), 212-224.

Delabays, N., Simonnet, X., & Gaudin, M. (2001). The genetics of artemisinin content in Artemisia annua L. and the breeding of high yielding cultivars. Curr Med Chem, 8(15), 1795-1801.

Desjardins, A. E., Hohn, T. M., & McCormick, S. P. (1993). Trichothecene biosynthesis in Fusarium species: chemistry, genetics, and significance. Microbiol Rev, 57(3), 595-604.

Dhingra, V., & Narasu, M. L. (2001). Purification and characterization of an enzyme involved in biochemical transformation of arteannuin B to artemisinin from Artemisia annua. Biochem Biophys Res Commun, 281(2), 558-561.

Dhingra, V., Vishweshwar Rao, K., & Lakshmi Narasu, M. (2000). Current status of artemisinin and its derivatives as antimalarial drugs. Life Sci, 66(4), 279-300.

Disbrow, G. L., Baege, A. C., Kierpiec, K. A., Yuan, H., Centeno, J. A., Thibodeaux, C. A., et al. (2005). Dihydroartemisinin Is Cytotoxic to Papillomavirus-Expressing Epithelial Cells In vitro and In vivo. Cancer Res, 65(23), 10854-10861.

Drew, M. G., Metcalfe, J., Dascombe, M. J., & Ismail, F. M. (2006). Reactions of artemisinin and arteether with acid: implications for stability and mode of antimalarial action. JMed Chem, 49(20), 6065-6073.

Duc, D. D., de Vries, P. J., Nguyen, X. K., Le Nguyen, B., Kager, P. A., & van Boxtel, C. J. (1994). The pharmacokinetics of a single dose of artemisinin in healthy Vietnamese subjects. Am J Trop Med Hyg, 51(6), 785-790.

Efferth, T. (2006). Molecular pharmacology and pharmacogenomics of artemisinin and its derivatives in cancer cells. Curr Drug Targets, 7(4), 407-421.

Efferth, T., Olbrich, A., & Bauer, R. (2002). mRNA expression profiles for the response of human tumor cell lines to the antimalarial drugs artesunate, arteether, and artemether. Biochemical Pharmacology, 64(4), 617-623.


63

Farmer, W. R., & Liao, J. C. (2001). Precursor balancing for metabolic engineering of lycopene production in Escherichia coli. Biotechnol Prog, 17(1), 57-61.

Fattorusso, C., Campiani, G., Catalanotti, B., Persico, M., Basilico, N., Parapini, S., et al. (2006). Endoperoxide Derivatives from Marine Organisms: 1,2-Dioxanes of the Plakortin Family as Novel Antimalarial Agents. J. Med. Chem., 49(24), 7088-7094.

Faulkner, A., Chen, X., Rush, J., Horazdovsky, B., Waechter, C. J., Carman, G. M., et al. (1999). The LPP1 and DPP1 gene products account for most of the isoprenoid phosphate phosphatase activities in Saccharomyces cerevisiae. J Biol Chem, 274(21), 14831-14837.

Ferreira, J. F. (2007). Nutrient deficiency in the production of artemisinin, dihydroartemisinic acid, and artemisinic acid in Artemisia annua L. J Agric Food Chem, 55(5), 1686-1694.

Ferreira, J. F., & Janick, J. (1996). Immunoquantitative analysis of artemisinin from Artemisia annua using polyclonal antibodies. Phytochemistry, 41(1), 97-104.

Ferreira, J. F., Simon, J. E., & Janick, J. (1995). Relationship of Artemisinin Content of Tissue-Cultured, Greenhouse-Grown, and Field-Grown Plants of Artemisia annua1. PlantaMed, 61(4), 351-355.

Flores, H. E., Vivanco, J. M., & Loyola-Vargas, V. M. (1999). 'Radicle' biochemistry: the biology of root-specific metabolism. Trends Plant Sci, 4(6), 220-226.

Gardner, R. G., & Hampton, R. Y. (1999). A highly conserved signal controls degradation of 3-hydroxy-3-methylglutaryl-coenzyme A (HMG-CoA) reductase in eukaryotes. J Biol Chem, 274(44), 31671-31678.

Greenhagen, B. T., O'Maille, P. E., Noel, J. P., & Chappell, J. (2006). Identifying and manipulating structural determinates linking catalytic specificities in terpene synthases. Proc Natl Acad Sci U S A, 103(26), 9826-9831.

Chapter 2

64

Han, J.-L., Liu, B.-Y., Ye, H.-C., Wang, H., Li, Z.-Q., & Li, G.-F. (2006). Effects of Overexpression of the Endogenous Farnesyl Diphosphate Synthase on the Artemisinin Content in Artemisia annua L. Journal of Integrative Plant Biology, 48(4), 482-487.

Henry, M., Alibert, S., Orlandi-Pradines, E., Bogreau, H., Fusai, T., Rogier, C., et al. (2006). Chloroquine Resistance Reversal Agents as Promising Antimalarial Drugs. Current Drug Targets, 7(8), 935-948.

Hohn, T. M., & Beremand, P. D. (1989). Isolation and nucleotide sequence of a sesquiterpene cyclase gene from the trichothecene-producing fungus Fusarium sporotrichioides. Gene, 79(1), 131-138.

Hoppe, H. C., van Schalkwyk, D. A., Wiehart, U. I. M., Meredith, S. A., Egan, J., & Weber, B. W. (2004). Antimalarial Quinolines and Artemisinin Inhibit Endocytosis in Plasmodium falciparum. Antimicrob. Agents Chemother., 48(7), 2370-2378.

Jackson, B. E., Hart-Wells, E. A., & Matsuda, S. P. (2003). Metabolic engineering to produce sesquiterpenes in yeast. Org Lett, 5(10), 1629-1632.

Jian Wen, W., & Ren Xiang, T. (2002). Artemisinin production in Artemisia annua hairy root cultures with improved growth by altering the nitrogen source in the medium. Biotechnology Letters, 24(14), 1153-1156.

Jung, M., Kim, H., Lee, K., & Park, M. (2003). Naturally occurring peroxides with biological activities. Mini Rev Med Chem, 3(2), 159-165.

Kaddouri, H., Nakache, S., Houze, S., Mentre, F., & Le Bras, J. (2006). Assessment of the drug susceptibility of Plasmodium falciparum clinical isolates from africa by using a Plasmodium lactate dehydrogenase immunodetection assay and an inhibitory maximum effect model for precise measurement of the 50-percent inhibitory concentration. Antimicrob Agents Chemother, 50(10), 3343-3349.


65

Kajiwara, S., Fraser, P. D., Kondo, K., & Misawa, N. (1997). Expression of an exogenous isopentenyl diphosphate isomerase gene enhances isoprenoid biosynthesis in Escherichia coli. Biochem J, 324 ( Pt 2), 421-426.

Kang, M. J., Lee, Y. M., Yoon, S. H., Kim, J. H., Ock, S. W., Jung, K. H., et al. (2005). Identification of genes affecting lycopene accumulation in Escherichia coli using a shot-gun method. Biotechnol Bioeng, 91(5), 636-642.

Kim, S. H., Heo, K., Chang, Y. J., Park, S. H., Rhee, S. K., & Kim, S. U. (2006). Cyclization mechanism of amorpha-4,11-diene synthase, a key enzyme in artemisinin biosynthesis. J Nat Prod, 69(5), 758-762.

Kim, S. W., & Keasling, J. D. (2001). Metabolic engineering of the nonmevalonate isopentenyl diphosphate synthesis pathway in Escherichia coli enhances lycopene production. Biotechnol Bioeng, 72(4), 408-415.

Klayman, D. L., Lin, A. J., Acton, N., Scovill, J. P., Hoch, J. M., Milhous, W. K., et al. (1984). Isolation of artemisinin (qinghaosu) from Artemisia annua growing in the United States. Journal of natural products, 47(4), 715-717.

Krishna, S., Uhlemann, A.-C., & Haynes, R. K. (2004). Artemisinins: mechanisms of action and potential for resistance. Drug Resistance Updates, 7(4-5), 233-244.

Kumar, S., Gupta, S. K., Singh, P., Bajpai, P., Gupta, M. M., Singh, D., et al. (2004). High yields of artemisinin by multi-harvest of Artemisia annua crops. Industrial Crops and Products 19(1), 77-90.

Lapkin, A. A., Plucinski, P. K., & Cutler, M. (2006). Comparative assessment of technologies for extraction of artemisinin. JNat Prod, 69(11), 1653-1664.

Laughlin, J. C. (2002). Post-harvest drying treatment effects on antimalarial constituents of Artemisia annua L. Acta Horticulturae, 576, 315-320.

Lesburg, C. A., Zhai, G., Cane, D. E., & Christianson, D. W. (1997). Crystal structure of pentalenene synthase: mechanistic

Chapter 2

66

insights on terpenoid cyclization reactions in biology. Science, 277(5333), 1820-1824.

Lindahl, A. L., Olsson, M. E., Mercke, P., Tollbom, O., Schelin, J., Brodelius, M., et al. (2006). Production of the artemisinin precursor amorpha-4,11-diene by engineered Saccharomyces cerevisiae. Biotechnol Lett, 28(8), 571-580.

Liu, C., Wang, Y., Guo, C., Ouyang, F., Ye, H., & Li, G. (1998). Enhanced production of artemisinin by Artemisia annua L hairy root cultures in a modified inner-loop airlift bioreactor. Bioprocess Engineering, 19(5), 389-392.

Liu, C. Z., Wang, Y. C., Ouyang, F., Ye, H. C., & Li, G. F. (1997). Production of artemisinin by hairy root cultures of Artemisia annua L. Biotechnology Letters, 19(9), 927-929.

Liu, Y., Wang, H., Ye, H.-C., & Li, G.-F. (2005). Advances in the Plant Isoprenoid Biosynthesis Pathway and Its Metabolic Engineering. Journal of Integrative Plant Biology, 47(7), 769-782.

Liu, Y., Wong, V. K. W., Ko, B. C. B., Wong, M. K., & Che, C. M. (2005). Synthesis and Cytotoxicity Studies of Artemisinin Derivatives Containing Lipophilic Alkyl Carbon Chains. Org. Lett., 7(8), 1561-1564.

Lommen, W. J., Schenk, E., Bouwmeester, H. J., & Verstappen, F. W. (2006). Trichome dynamics and artemisinin accumulation during development and senescence of Artemisia annua leaves. Planta Med, 72(4), 336-345.

Loreti, E., Alpi, A., & Perata, P. (2000). Glucose and disaccharide-sensing mechanisms modulate the expression of alpha-amylase in barley embryos. Plant Physiol, 123(3), 939-948.

Ma, C., Wang, H., Lu, X., Li, H., Liu, B., & Xu, G. (2007). Analysis of Artemisia annua L. volatile oil by comprehensive two-dimensional gas chromatography time-of-flight mass spectrometry. Journal of Chromatography A, 1150(1-2), 50-53.

Martin, V. J., Pitera, D. J., Withers, S. T., Newman, J. D., & Keasling, J. D. (2003). Engineering a mevalonate pathway in


67

Escherichia coli for production of terpenoids. NatBiotechnol, 21(7), 796-802.

Math, S. K., Hearst, J. E., & Poulter, C. D. (1992). The crtE gene in Erwinia herbicola encodes geranylgeranyl diphosphate synthase. Proc Natl Acad Sci U S A, 89(15), 6761-6764.

Merali, S., & Meshnick, S. R. (1991). Susceptibility of Pneumocystis carinii to artemisinin in vitro. Antimicrob Agents Chemother, 35(6), 1225-1227.


Mercke, P., Crock, J., Croteau, R., & Brodelius, P. E. (1999). Cloning, expression, and characterization of epi-cedrol synthase, a sesquiterpene cyclase from Artemisia annua L. Arch Biochem Biophys, 369(2), 213-222.

Messori, L., Piccioli, F., Temperini, C., Bilia, A. R., Vincieri, F. F., Allegrozzi, M., et al. (2004). The reaction of artemisinin with hemin: a further insight into the mechanism. Inorganica Chimica Acta, 357(15), 4602-4606.

Mueller, M. S., Karhagomba, I. B., Hirt, H. M., & Wemakor, E. (2000). The potential of Artemisia annua L. as a locally produced remedy for malaria in the tropics: agricultural, chemical and clinical aspects. Journal of Ethnopharmacology, 73(3), 487-493.

Mueller, M. S., Runyambo, N., Wagner, I., Borrmann, S., Dietz, K., & Heide, L. (2004). Randomized controlled trial of a traditional preparation of Artemisia annua L. (Annual Wormwood) in the treatment of malaria. Trans R Soc Trop Med Hyg, 98(5), 318-321.

Newman, J. D., Marshall, J., Chang, M., Nowroozi, F., Paradise, E., Pitera, D., et al. (2006). High-level production of amorpha-4,11-diene in a two-phase partitioning bioreactor of

Chapter 2

68

metabolically engineered Escherichia coli. Biotechnol Bioeng, 95(4), 684-691.

O'Neill, P. M., Rawe, S. L., Borstnik, K., Miller, A., Ward, S. A., Bray, P. G., et al. (2005). Enantiomeric 1,2,4-Trioxanes Display Equivalent in vitro Antimalarial Activity Versus Plasmodium falciparum Malaria Parasites: Implications for the Molecular Mechanism of Action of the Artemisinins. ChemBioChem, 6(11), 2048-2054.

Panossian, L. A., Garga, N. I., & Pelletier, D. (2005). Toxic brainstem encephalopathy after artemisinin treatment for breast cancer. Ann Neurol, 58(5), 812-813.

Peng, C. A., Ferreira, J. F., & Wood, A. J. (2006). Direct analysis of artemisinin from Artemisia annua L. using high-performance liquid chromatography with evaporative light scattering detector, and gas chromatography with flame ionization detector. J Chromatogr A, 1133(1-2), 254-258.

Peres, V. F., Saffi, J., Melecchi, M. I. S., Abad, F. C., de Assis Jacques, R., Martinez, M. M., et al. (2006). Comparison of soxhlet, ultrasound-assisted and pressurized liquid extraction of terpenes, fatty acids and Vitamin E from Piper gaudichaudianum Kunth. Journal of Chromatography A, 1105(1-2), 115-118.

Pfleger, B. F., Pitera, D. J., Smolke, C. D., & Keasling, J. D. (2006). Combinatorial engineering of intergenic regions in operons tunes expression of multiple genes. Nat Biotechnol, 24(8), 1027-1032.


Picaud, S., Olofsson, L., Brodelius, M., & Brodelius, P. E. (2005). Expression, purification, and characterization of recombinant amorpha-4, 11-diene synthase from Artemisia annua L. Archives of Biochemistry and Biophysics, 436(2), 215-226.


69

Ploypradith, P. (2004). Development of artemisinin and its structurally simplified trioxane derivatives as antimalarial drugs. Acta Trop, 89(3), 329-342.

Posner, G. H., & O'Neill, P. M. (2004). Knowledge of the Proposed Chemical Mechanism of Action and Cytochrome P450 Metabolism of Antimalarial Trioxanes Like Artemisinin Allows Rational Design of New Antimalarial Peroxides. Acc. Chem. Res., 37(6), 397-404.

Proctor, R. H., & Hohn, T. M. (1993). Aristolochene synthase. Isolation, characterization, and bacterial expression of a sesquiterpenoid biosynthetic gene (Ari1) from Penicillium roqueforti. J Biol Chem, 268(6), 4543-4548.

Rafiee, M. A., Hadipour, N. L., & Naderi-manesh, H. (2005). The role of charge distribution on the antimalarial activity of artemisinin analogues. J Chem Inf Model, 45(2), 366-370.

Ram, M., Gupta, M. M., Dwivedi, S., & Kumar, S. (1997). Effect of Plant Density on the Yields of Artemisinin and Essential Oil in Artemisia annua Cropped under Low Input Cost Management in North-Central India. Planta Med, 63(4), 372-374.

Ram, M., Gupta, M. M., Naqvi, A. A., & Kumar, S. (1997). Effect of planting time on the yield of essential oil and artemisinin in Artemisia annua under subtropical conditions. Journal of Essential Oil Research, 9, (2), 193-197.

Rath, K., Taxis, K., Walz, G., Gleiter, C. H., Li, S. M., & Heide, L. (2004). Pharmacokinetic study of artemisinin after oral intake of a traditional preparation of Artemisia annua L. (annual wormwood). Am J Trop Med Hyg, 70(2), 128-132.


Rohdich, F., Hecht, S., Gartner, K., Adam, P., Krieger, C., Amslinger, S., et al. (2002). Studies on the nonmevalonate

Chapter 2

70

terpene biosynthetic pathway: metabolic role of IspH (LytB) protein. Proc Natl Acad Sci U S A, 99(3), 1158-1163.

Rolland, F., Baena-Gonzalez, E., & Sheen, J. (2006). Sugar sensing and signaling in plants: conserved and novel mechanisms. Annu Rev Plant Biol, 57, 675-709.

Romero, M. R., Serrano, M. A., Vallejo, M., Efferth, T., Alvarez, M., & Marin, J. J. (2006). Antiviral effect of artemisinin from Artemisia annua against a model member of the Flaviviridae family, the bovine viral diarrhoea virus (BVDV). Planta Med, 72(13), 1169-1174.

Roth, R. J., & Acton, N. (1989). A simple conversion of artemisinic acid into artemisinin. J Nat Prod, 52(5), 1183-1185.

Rynkiewicz, M. J., Cane, D. E., & Christianson, D. W. (2001). Structure of trichodiene synthase from Fusarium sporotrichioides provides mechanistic inferences on the terpene cyclization cascade. Proc Natl Acad Sci U S A, 98(24), 13543-13548.

Sa, G., Mi, M., He-chun, Y., Ben-ye, L., Guo-feng, L., & Kang, C. (2001). Effects of ipt gene expression on the physiological and chemical characteristics of Artemisia annua L. Plant Sci, 160(4), 691-698.

Sangwan, R. S., Agarwal, K., Luthra, R., Thakur, R. S., & Singh-Sangwan, N. (1993). Biotransformation of arteannuic acid into arteannuin-B and artemisinin in Artemisia annua. Phytochemistry, 34(5), 1301-1302.

Schmuck, G., Roehrdanz, E., Haynes, R. K., & Kahl, R. (2002). Neurotoxic Mode of Action of Artemisinin. Antimicrob.Agents Chemother., 46(3), 821-827.

Schnee, C., Kollner, T. G., Gershenzon, J., & Degenhardt, J. (2002). The maize gene terpene synthase 1 encodes a sesquiterpene synthase catalyzing the formation of (E)-beta-farnesene, (E)-nerolidol, and (E,E)-farnesol after herbivore damage. Plant Physiol, 130(4), 2049-2060.

Shanks, J. V., & Morgan, J. (1999). Plant 'hairy root' culture. CurrOpin Biotechnol, 10(2), 151-155.


71

Sisowath, C., Stromberg, J., Martensson, A., Msellem, M., Obondo, C., BjÃ¶rkman, A., et al. (2005). In Vivo Selection of Plasmodium falciparum pfmdr1 86N Coding Alleles by Artemether-Lumefantrine (Coartem). Journal of Infectious Diseases, 191(6), 1014-1017.

Souret, F. F., Kim, Y., Wyslouzil, B. E., Wobbe, K. K., & Weathers, P. J. (2003). Scale-up of Artemisia annua L. hairy root cultures produces complex patterns of terpenoid gene expression. Biotechnol Bioeng, 83(6), 653-667.

Starks, C. M., Back, K., Chappell, J., & Noel, J. P. (1997). Structural basis for cyclic terpene biosynthesis by tobacco 5-epi-aristolochene synthase. Science, 277(5333), 1815-1820.

Sy, L.-K., & Brown, G. D. (2002). The role of the 12-carboxylic acid group in the spontaneous autoxidation of dihydroartemisinic acid. Tetrahedron, 58(5), 909-923.

Sy L.-K., & Brown., G. D. . (2002). The mechanism of the spontaneous autoxidation of dihydroartemisinic acid. Tetrahedron, 58, 897-908.

Takahashi, S., Yeo, Y., Greenhagen, B. T., McMullin, T., Song, L., Maurina-Brunker, J., et al. (2006). Metabolic engineering of sesquiterpene metabolism in yeast. Biotechnol Bioeng.

Taylor, W. R. J., & White, N. J. (2004). Antimalarial Drug Toxicity. Drug Safety, 27(1), 25-61.

Teoh, K. H., Polichuk, D. R., Reed, D. W., Nowak, G., & Covello, P. S. (2006). Artemisia annua L. (Asteraceae) trichome-specific cDNAs reveal CYP71AV1, a cytochrome P450 with a key role in the biosynthesis of the antimalarial sesquiterpene lactone artemisinin. FEBS Lett, 580(5), 1411-1416.

Titulaer, H. A., Zuidema, J., Kager, P. A., Wetsteyn, J. C., Lugt, C. B., & Merkus, F. W. (1990). The pharmacokinetics of artemisinin after oral, intramuscular and rectal administration to volunteers. J Pharm Pharmacol, 42(11),810-813.

Urban, P., Mignotte, C., Kazmaier, M., Delorme, F., & Pompon, D. (1997). Cloning, yeast expression, and characterization of

Chapter 2

72

the coupling of two distantly related Arabidopsis thaliana NADPH-cytochrome P450 reductases with P450 CYP73A5. J Biol Chem, 272(31), 19176-19186.

van Agtmael, M. A., Eggelte, T. A., & van Boxtel, C. J. (1999). Artemisinin drugs in the treatment of malaria: from medicinal herb to registered medication. Trends Pharmacol Sci, 20(5), 199-205.

Van Nieuwerburgh, F. C., Vande Casteele, S. R., Maes, L., Goossens, A., Inze, D., Van Bocxlaer, J., et al. (2006). Quantitation of artemisinin and its biosynthetic precursors in Artemisia annua L. by high performance liquid chromatography-electrospray quadrupole time-of-flight tandem mass spectrometry. J Chromatogr A, 1118(2), 180-187.

Vedula, L. S., Rynkiewicz, M. J., Pyun, H. J., Coates, R. M., Cane, D. E., & Christianson, D. W. (2005). Molecular recognition of the substrate diphosphate group governs product diversity in trichodiene synthase mutants. Biochemistry, 44(16), 6153-6163.

Vogeli, U., Freeman, J. W., & Chappell, J. (1990). Purification and Characterization of an Inducible Sesquiterpene Cyclase from Elicitor-Treated Tobacco Cell Suspension Cultures. Plant Physiol, 93(1), 182-187.


Wallaart, T. E., Pras, N., Beekman, A. C., & Quax, W. J. (2000). Seasonal variation of artemisinin and its biosynthetic precursors in plants of Artemisia annua of different geographical origin: proof for the existence of chemotypes. Planta Med, 66(1), 57-62.

Wallaart, T. E., Pras, N., & Quax, W. (1999). Seasonal Variations of Artemisinin and its Biosynthetic Precursors in Tetraploid


73

Artemisia annua Plants Compared with the Diploid Wild-Type. Planta Medica(8), 723-728.

Wallaart, T. E., Pras, N., & Quax, W. J. (1999). Isolation and identification of dihydroartemisinic acid hydroperoxide from artemisia annua: A novel biosynthetic precursor of artemisinin. J Nat Prod, 62(8), 1160-1162.

Wallaart, T. E., van Uden, W., Lubberink, H. G., Woerdenbag, H. J., Pras, N., & Quax, W. J. (1999). Isolation and identification of dihydroartemisinic acid from artemisia annua and its possible role in the biosynthesis of artemisinin. J Nat Prod, 62(3), 430-433.

Wang, H., Ge, L., Ye, H. C., Chong, K., Liu, B. Y., & Li, G. F. (2004). Studies on the effects of fpf1 gene on Artemisia annua flowering time and on the linkage between flowering and artemisinin biosynthesis. Planta Med, 70(4), 347-352.

Wang, H., Liu, Y., Chong, K., Liu, B. Y., Ye, H. C., Li, Z. Q., et al. (2006). Earlier Flowering Induced by Over-Expression of CO Gene Does Not Accompany Increase of Artemisinin Biosynthesis in Artemisia annua. Plant Biol (Stuttg).

Wang, J., Xia, Z., & Tan, R. (2002). Elicitation on artemisinin biosynthesis in Artemisia annua hairy roots by oligosaccharide extract from endophytic Colletotrichum sp. B501. Acta Botanica Sinica, 44(10), 1233-1238.

Wang, M., Park, C., Wu, Q., & Simon, J. E. (2005). Analysis of artemisinin in Artemisia annua L. by LC-MS with selected ion monitoring. J Agric Food Chem, 53(18), 7010-7013.

Weathers, P. J., DeJesus-Gonzalez, L., Kim, Y. J., Souret, F. F., & Towler, M. J. (2004). Alteration of biomass and artemisinin production in Artemisia annua hairy roots by media sterilization method and sugars. Plant Cell Rep, 23(6), 414-418.

Weathers, P. J., Hemmavanh, D. D., Walcerz, D. B., Cheetham, R. D., & Smith, T. C. (1997). Interactive effects of nitrate and phosphate salts, sucrose, and inoculum culture age on growth and sesquiterpene production in Artemisia annua

Chapter 2

74

hairy root cultures. In Vitro Cellular & Developmental Biology - Plant 33(4), 306-312.

White, N. J. (2004). Antimalarial drug resistance. J. Clin. Invest., 113(8), 1084-1092.

Woerdenbag, H. J., Lugt, C. B., & Pras, N. (1990). Artemisia annua L.: a source of novel antimalarial drugs. Pharm Weekbl Sci, 12(5), 169-181.

Woerdenbag, H. J., Pras, N., van Uden, W., Wallaart, T. E., Beekman, A. C., & Lugt, C. B. (1994). Progress in the research of artemisinin-related antimalarials: an update. Pharm World Sci, 16(4), 169-180.

Wong, J. W., & Yuen, K. H. (2001). Improved oral bioavailability of artemisinin through inclusion complexation with beta- and gamma-cyclodextrins. Int J Pharm, 227(1-2), 177-185.

Wong, J. W., & Yuen, K. H. (2003). Inclusion complexation of artemisinin with alpha-, beta-, and gamma-cyclodextrins. Drug Dev Ind Pharm, 29(9), 1035-1044.

Yadav, J. S., Satheesh Babu, R., & Sabitha, G. (2003). Stereoselective total synthesis of (+)-artemisinin. Tetrahedron Letters, 44(2), 387-389.

Zhou, W.-l., Wu, J.-m., Wu, Q.-l., Wang, J.-x., Zhou, Y., Zhou, R., et al. (2005). A novel artemisinin derivative, 3-(12-beta-artemisininoxy) phenoxyl succinic acid (SM735), mediates immunosuppressive effects in vitro and in vivo1. Acta Pharmacologica Sinica, 26(11), 1352-1358.

Ziffer, H., Highet, R. J., & Klayman, D. L. (1997). Artemisinin: an endoperoxidic antimalarial from Artemisia annua L. Fortschr Chem Org Naturst, 72, 121-214.

Chapter 3

Molecular cloning and characterization of a broad

substrate terpenoid oxidoreductase from Artemisia annua

Anna-Margareta Rydén1, Carolien Ruyter-Spira2, Ralph Litjens2,Shunji Takahashi3, Wim Quax1, Hiroyuki Osada3, Harro Bouwmeester4 and Oliver Kayser1

1Department of Pharmaceutical Biology, GUIDE, University of Groningen, 9713 AV Groningen, The Netherlands 2Metabolic Regulation Division, Plant Research International, Wageningen 6700 AA, the Netherlands, The Netherlands 3Department of Chemical Biology, RIKEN, Saitama 356-0004, Japan 4Laboratory of Plant Physiology, Wageningen University, P.O. Box 658, 6700 AR Wageningen, the Netherlands

Adapted from: Plant and Cell Physiology (2010). In press.

Chapter 3

76

Abstract

From Artemisia annua L., a new oxidoreductase (Red 1) was cloned, sequenced and functionally characterized. Through bioinformatics, heterologous protein expression, and enzyme substrate conversion assays, the elucidation of the enzymatic capacities of Red1 was achieved. Red1 acts on monoterpenoids, and in particular functions as a menthone:neomenthol oxidoreductase. The kinetic parameter kcat/Km was determined to be 939 fold more efficient for the reduction of (-)-menthone to (+)-neomenthol, than results previously reported for the menthone:neomenthol reductase from Mentha x piperita. Based on its kinetic properties, the possible use of Red1 in biological crop protection is discussed. .

Artemisia annua terpenoid oxidoreductase

77

Introduction

Artemisia annua L. (A. annua) is primarily known for its production of the antimalarial drug artemisinin, which is extracted from the plant in small quantities (0.01-2% dry weight (Zhang et al., 2008)). Artemisinin belongs to the sesquiterpenoids, and this is also the class of compounds that has attracted most attention from researchers investigating this plant (Bertea et al., 2006; Bouwmeester et al., 1999; Mercke et al., 2000; Wallaart et al., 1999; Zhang et al., 2008). Less investigated is the biosynthesis of its monoterpenoid metabolic profile, even though studies of the composition of A. annua essentialoil show that it contains many different monoterpenoids (Goelet al., 2007; Woerdenbag et al., 1993; Woerdenbag et al., 1994). Hence, in addition to artemisinin, A. annua contains many other terpenoids of commercial interest such as menthol (table 1). The enantiomer (-)-menthol is used in industry and pharmacy and is recognized as a high value natural compound. For biotechnological purposes, it would be desirable to isolate enzymes capable of producing menthol or other closely related compounds, such as (+)-neomenthol. In order to begin to understand the origin of oxygenated terpenoids in A. annua, we cloned and characterized an oxidoreductase, Red1 (Rydén et al., 2010). We showed that Red1 is very likely to be involved in the biosynthetic network of artemisinin. To investigate the potential use of Red1 in biotechnological applications, we here investigate the substrate specificity of the enzyme in more detail.

Previously, EST libraries from organs specialized in terpenoid biosynthesis have been used to clone and characterize enzymes involved in artemisinin biosynthesis (Covello et al., 2007; Mercke et al., 2000; Ro et al., 2006; Wallaart et al., 2001; Zhang et al., 2008). Here, we describe the heterologous over-expression and isolation of Red1, and investigate its enzymatic properties toward monoterpenoids.

Chapter 3

78

Table 1. Phytochemical profile of A. annua.

Substrate

% of substrate

in essential oil

Convertedby Red1 References

Artemisinic alcohol 5.2-7.5 - (Woerdenbag et al., 1993)


77-24501) - (Lommen et al., 2006)

Borneol 0.6-20.0 - (Woerdenbag et al., 1994), (Banthorpe et al., 1977)

Camphor 3.3-21.8 - (Woerdenbag et al., 1993), (Bhakuni et al., 2001)

Carvone Traces - (Goel et al., 2007)

Citral n.d. -


35-19841) - (Lommen et al., 2006)


79

Table 1. Continuation.

Substrate

% of substrate

in essential oil



23-7811) + (Lommen et al., 2006), (Rydén et al., 2010)

Dihydrocarveol n.d. -

Dihydrocarvone n.d. +

Germacrone n.d. -

Humulene 0.2-0.7 - (Woerdenbag et al., 1994)

Isopiperitenol n.d. -

Limonene 0.23 - (Ahmad et al., 1994)

Menthol and

derivatives

Minorcomponents

+ (Chalchat et al., 1991), (Lawrence, 1982)

Menthone n.d. +

Neomenthol n.d. +

Nootkatone n.d. -

Perilla alcohol n.d. -

Perilla aldehyde n.d. + (Rydén et al., 2010)

Chapter 3

80

Table 1. Continuation.

Substrate

% of substrate

in essential oil


Terpineol 0.1-0.9 - (H. J. Woerdenbag, et al., 1994)

Thujone Traces-9.0 + (Derek V. Banthorpe, et al., 1977), (D. V. Banthorpe, Baxendale, Gatford, & Williams, 1971), (Ahmad & Misra, 1994)

1) μg g-1 dry weight plant material. n.d. – no data available.

Incubations of Red1 with a series of monoterpenoids such as menthone, carvone and limonene as well as sesquiterpenoids exemplified by nootkatone and germacrone, were performed to investigate whether Red1 is capable of reducing the ketone group to an alcohol in these systems, along with the reverse dehydrogenase reaction (figure 1). The contribution of Red1 to the metabolic profile of A. annua is discussed, as well as the prospects of Red1 in transgenic approaches.


81

a

OH

(+)-neomenthol

O

(-)-menthone

OH

(-)-menthol

O

(+)-carvone

O

(-)-dihydrocarvone

OH

(+)-dihydrocarveol

O

(-)-carvone

O

(+)-dihydrocarvone

OH

(+)-dihydrocarveol

O

beta-thujone

O

alpha-thujone

OH

thujanol

Chapter 3

82

b

OH

(-)-borneol

OH

(+)-borneol

O

(+)-camphor

O

(-)-camphor (-)-limonene

OH

isopiperitenol

O

(+)-menthone

OH

(+)-menthol

OH

(-)-neomentholOH

1-alpha-terpineol

Ocitral

O

nootkatone

O

germacrone alpha-humulene

Figure 1. Substrates and conversion products for Red1. Solid arrows reprecent conversion by Red1. a. Dashed arrow represents conversion via innate E. coliproteins.

b. Substrates not converted by Red1.

Results and Discussion

Sequence analysis

After sequencing 2180 ESTs and running a batch BlastX, 19 reductase candidates were identified and assigned putative functions. To further narrow down the number of candidates, a comparison with tomato, potato, rice, Arabidopsis thaliana (A. thailana), lettuce and sunflower EST libraries was done. To distinguish between


83

ubiquitous reductases and reductases likely to be relevant in sesquiterpene lactone biosynthesis, the following strategy was adopted: Reductases with a high similarity with lettuce and sunflower, but significantly lower similarity with the non-Asteraceae outgroups potato (Solanaceae), tomato (Solanaceae), rice (Poaceae)and A. thaliana (Brassicaceae), were considered to be promising candidates in the terpenoid biosynthesis. Based on the results from the BlastX search and the EST library comparisons, one candidate named red1 was selected as the most promising candidate to relevant terpenoid biosynthesis. This was inferred from its similarity with the monoterpenoid reductase isopiperitenone reductase (Ringer et al., 2003) and its high similarity with a lettuce (74%) and sunflower EST (43%). The comparatively low similarity with rice (24%) and A.thaliana (33%) further solidified the hypothesis of Red1 to be a promising candidate. The full length sequence was therefore obtained using RACE-PCR.

Analysis of the full length sequence of red1 with tBlastX revealed Mentha x piperita (M. piperita) (-)-isopiperitenone reductase, M.piperita (-)-menthone:(+)-neomenthol reductase, M. piperita menthol dehydrogenase and Papaver bracteatum (P. bracteatum) salutaridine reductase as top hits, with enzymatically confirmed functions. Alignment of these amino acid sequences including the deduced amino acid sequence of Red1 illustrate two conserved motifs, which are common for the short chain dehydrogenase/reductase (SDR) protein family (figure 2) (Kallberg et al., 2002): the N-terminal cofactor binding motif GxxxGxG (motif 1) and the catalytic domain YxxxK (motif 2). Common to the enzymes is their preference for NADP(H) over NAD(H), due to the basic lysine residue in motif 1 (Persson et al., 2003), which categorizes them as classical SDR enzymes (Kallberg et al., 2002). Two-by-two comparison of amino acid sequences produced similarity rankings of Red1 aligned with P.bracteatum salutaridine reductase (49% identity), M. piperita (-)-isopiperitenone reductase and M. piperita menthol dehydrogenase

Chapter 3

84

(47% identity) and M. piperita (-)-menthone:(+)-neomenthol reductase (46% identity).

Motif 1MNR MGDEVVVDHAATKRYAVVT GANKGIG FEICKQLASKGITVILASRDEKRGIEARERLIKE 60MOHDH ------MADTFTQRYALVT GANKGIG FEICRQLASKGMKVILASRNEKRGIEARERLLKE 54MIR --------MAEVQRYALVT GANKGIG FEICRQLAEKGIIVILTSRNEKRGLEARQKLLKE 52SR MP-ETCPNTVTKMRCAVVT GGNKGIG FEICKQLSSSGIMVVLTCRDVTRGLEAVEKLK-- 57red1 MSYATEKDVSTEKRVALVT GGNKGIG LEICRQLASNDIKVILTARNESRGIEAIEKLK-- 58

* *:** *.***** :***:**:...: *:*:.*: .**:** ::*

MNR LGSEFGDYVVSQQLDVADP-ASVAALVDFIKTKFGSLDILVNNAGLNGTYMEGDASVLND 119MOHDH SRSISDDDVVFHQLDVADP-ASAVAVAHFIETKFGRLDILVNNAGFTGVAIEGDISVYQE 113MIR LN-VSENRLVFHQLDVTDL-ASVAAVAVFIKSKFGKLDILVNNAGVSGVEMVGDVSVFNE 110SR --NSNHENVVFHQLDVTDPITTMSSLADFIKARFGKLDILVNNAGVAGFSVDADR--FKA 113red1 --VSGPLDVVFHQLDVKDP-SSIARLAKYVELQFKKLDILVNNAGESGIIVREDE--FRA 113

:* :**** * :: :. ::: :* ********* * : * .

MNR YVEAEFKTFQ---SGAAKTEPYHPKATGRLVETVEHAKECIETNYYGSKRVTEALIPLLQ 176MOHDH CLEANIIAAQ---GGQAH--PFHPKTTGRLIETLEGSKECIETNYYGTKRITETLIPLLQ 168MIR YIEADFKALQALEAGAKEEPPFKPKANGEMIEKFEGAKDCVVTNYYGPKRLTQALIPLLQ 170SR MISDIGEDSE-----EVVKIYEKPEAQELMSETYELAEECLKINYYGVKSVTEVLLPLLQ 168red1 FKDGAGYN-------EVYDEN-AHLLTEIIEQPPHLGEECIKTNYYGTKGVTEAFLPLLQ 165

. : : . .::*: **** * :*:.::****

MNR QSDSPRIVNVSSTLSSLVFQTNEWAKGVFSSEEGLTEEKLEEVLAEFLKDFIDGKQQEKQ 236MOHDH KSDSPTIVNVSSTFSTLLLQPNEWAKGVFSSN-SLNEGKVEEVLHEFLKDFIDGKLQQNH 227MIR LSPSPRIVNVSSSFGSLLLLWNEWAKGVLGDEDRLTEERVDEVVEVFLKDIKEGKLEESQ 230SR LSDSPRIVNVSSSTGSLKYVSNETALEILGDGDALTEERIDMVVNMLLKDFKENLIETNG 228red1 LSKSLRIVNVSSNYGELKFLPNEKLTQELQDIEHLTNERIDEIIQWXLRDLKANKLLENG 225

* * ******. . * ** : . *.: ::: :: *:*: . . Motif 2

MNR WPPHFSA YKVSK AALNAYTRIIAKKYPSFRINAVCPGYTKTDLSYGHGQFTDAEAAEAPV 296MOHDH WPPNFAA YKVSK AAVNAYTRIIARKYPSFCINSVCPGFVRTDICYNLGVLSEAEGAEAPV 287MIR WPPHFAA ERVSK AALNAYTKIAAKKYPSFRINAICPGYAKTDITFHAGPLSVAEAAQVPV 290SR WPSFGAA YTTSK ACLNAYTRVLAKKIPKFQVNCVCPGLVKTEMNYGIGNYTADEGAKHVV 288red1 WPLTVGA YKISK IAVNAYTRLLARKYQNILVNCVHPGYVITDITSNTGELTSEEGAKAPV 285

** .* ** .:****:: *:* .: :*.: ** . *:: * : *.*: *

MNR KLALLPQGGPSGCFFFRDEAFCLY 320MOHDH KLALLPDGGPSGSFFSREEALSLY 311MIR KLALLPDGGPSGCFFPRDKALALY 314SR RIALFPDDGPSGFFYDCSELSAF- 311red1 MVALLPDDGPSGVYFSRMQITSF- 308

:**:*:.**** :: : .:

Figure 2. Alignment of homologous amino acid sequences against Red1. MIR = Mentha x piperita (-)-isopiperitenone reductase (accession number AY300162) (Ringer et al., 2003), MNR = Mentha x piperita (-)-menthone:(+)-neomenthol reductase (accession ABC88670) (Daviset al., 2005), MOHDH = Mentha x piperita menthol dehydrogenase (accession number AY288138) (Davis et al., 2005) and SR = Papaver bracteatumsalutaridine reductase (accession number EF184229) (Geissler et al., 2007). Motif 1 GxxxGxG is a cofactor binding domain typical for SDR (short chain dehydrogenase/reductase


85

superfamily) and motif 2 YxxxK is a catalytic domain (Davis et al., 2005; Kallberg et al., 2002). (*): strictly conserved residues; (:): conserved substitutions; (.): semiconserved substitutions.

Reductase activities

Based on the bioinformatics analysis, it was hypothesized that Red1 would reduce exocyclic double bonds ketones. Initial experiments were performed with crude cytosolic extracts from Escherichia coli(E. coli) producing Red1 without purification. The crude protein extract was tested for conversion of (-)-camphor, (+)-camphor, (+)-carvone, (-)-carvone, citral, germacrone, -humulene, (-)-limonene, (-)-menthone mixed with 10% (+)-menthone, (+)-nootkatone, 1- -terpineol and -thujone with NADPH as cofactor (figure 1). Conversion of the ketones into their corresponding alcohols was detected for the terpenes menthone, carvone and -thujone, but not for the other monoterpenes and sesquiterpene ketones. Menthone was predominantly converted into (-)-menthol, with traces of neomenthol (figure 3a, b). Both enantiomers of carvone were converted into dihydrocarveol, but biotransformation of carvone to dihydrocarvone also occurred in the negative control (data not shown). -Thujone was to a minor extent converted into its alcohol (figure 1, data not shown). This demonstrates that the reduction of the cyclic carbon double bond is due to native E. coli proteins. To discriminate between Red1 dependent conversion and conversion by E. coli enzymes, his-tagged Red1 protein was used in assays with selected monoterpenoids. With NADPH as cofactor, dihydrocarvone (figure 4a, b) and menthone (figure 3c, d, e) were efficiently converted to corresponding alcohols dihydrocarveol and menthol/neomenthol respectively (figure 1). The monoterpenes carvone and isopiperitenone were not converted (data not shown), suggesting that Red1 preferentially reduces saturated ketones. In a crude protein extract containing his-tagged Red1, (-)-menthone was converted predominantly to (-)-menthol with traces of

Chapter 3

86

a

0

10

20

30

40

50

Tot

al Io

n C

urre

nt

14.50 15.00Time (min)

1

2

3

4

c

0

10

20

30

40

50

60

70

80

90

100

Tot

al Io

n C

urre

nt

14.50 15.00Time (min)

1

2

3

4

b

0

10

20

30

40

50

60

70

80

Tot

al Io

n C

urre

nt

14.50 15.00Time (min)

1

2

d

0

10

20

30

40

50

60

70

80

90

100

Tot

al Io

n C

urre

nt

14.50 15.00Time (min)

1

2

e

0

10

20

Tot

al Io

n C

urre

nt

14.50 15.00Time (min)

1

2

3

4


87

Figure 3. GC chromatogram of conversion assays. 1 = (-)-menthone, 2 = (+)-menthone, 3 = neomenthol, 4 = (-)-menthol.

a. Crude preparation of Red1 with (-)-menthone and (+)-menthone as substrate. b. Crude empty vector control with (-)-menthone and (+)-menthone as substrate. c. Purified his-tagged Red1 with (-)-menthone and (+)-menthone as substrate. d. Empty vector control with (-)-menthone and (+)-menthone as substrate. e. Crude preparation of his-tagged Red1 with (-)-menthone and (+)-menthone as substrate.

neomenthol (figure 3e), whereas purified his-tagged Red1 converted (-)-menthone to (+)-neomenthol (main product) and (-)-menthol (figure 3c). The results with purified his-tagged Red1 confirmed the function of red1 to reduce ketone groups in saturated systems.

a

0

10

20

30

40

50

60

Tot

al Io

n C

urre

nt

15.40 15.60 15.80 16.00 16.20Time (min)

5

6

b

0102030405060708090

100

Tot

al Io

n C

urre

nt

15.40 15.60 15.80 16.00 16.20Time (min)

7

8

Figure 4. GC chromatogram of conversion assays. 5 = (-)-dihydrocarveol, 6 = (+)-dihydrocarveol, 7 = (-)-dihydrocarvone, 8 = (+)-dihydrocarvone.

a. Purified his-tagged Red1 with (-)-dihydrocarvone as substrate. b. Empty vector control with (-)-dihydrocarvone as substrate.

Chapter 3

88

Dehydrogenase activities

The capability of Red1 to perform the reverse reaction, oxidation of alcohol to ketone or aldehyde groups, was investigated using selected monoterpenes. Purified his-tagged Red1 was incubated with the monoterpenes (+)-borneol, (-)-borneol, (+)-menthol, (-)-menthol, (+)-neomenthol and (-)-neomenthol with NADP+ as cofactor (figure 1). The compounds (+)-borneol, (-)-borneol, (+)-menthol and (-)-neomenthol were not converted. (+)-Neomenthol proved to be well accepted and converted into (-)-menthone (figure 5a, b). Red1 showed interesting enantiomer specificity towards menthol, as it only accepted (-)-menthol as a substrate but not (+)-menthol (figure 6a, b). Another interesting property of Red1 is its ability to convert neomenthol to (-)-menthol via the intermediate (-)-menthone (figure 5a, b). These results show the capability of Red1 to reduce, as well as oxidize, monoterpenes. Of the monoterpenoid substrates tested, Red1 showed highest activity for the conversion of (-)-menthone to (+)-neomenthol.

a

0

10

20

30

40

50

60

Tot

al Io

n C

urre

nt

14.50 15.00Time (min)

1

3 4

b

0102030405060708090

100

Tot

al Io

n C

urre

nt

14.50 15.00Time (min)

3

Figure 5. GC chromatogram of conversion assays. 1 = (-)-menthone, 3 = neomenthol, 4 = (-)-menthol.

a. Purified his-tagged Red1 with (+)-neomenthol as substrate. b. Empty vector control with (+)-neomenthol as substrate.


89

a

0

10

20

30

40

50

Tot

al Io

n C

urre

nt

14.50 15.00Time (min)

1

4

b

0

10

20

30

40

50

Tot

al Io

n C

urre

nt

14.50 15.00Time (min)

4

Figure 6. GC chromatogram of conversion assays. 1 = (-)-menthone, 4 = (-)-menthol.

a. Purified his-tagged Red1 with (-)-menthol as substrate. b. Empty vector control with (-)-menthol as substrate.

Kinetic parameters

Activity of purified Red1 was studied in different buffers, by dialyzing against buffers A-F (table 2). The results show that Red1 is sensitive toward reducing agents, implying importance of sulphur bridges for activity, and further that the enzyme is stabilized by the addition of salt. The dependence of conversion activity on buffer pH was investigated, by comparing conversion assays performed in either potassium phosphate buffers, or Tris buffers, over a range of pH 6.0 to 9.0 with 0.5 unit increments. Red1 shows a broad pH optimum, ranging from 6.0 to 8.0 with maximum activity at pH 7.0. The optimal temperature for Red1 for menthone:neomenthol conversion was 30°C. The Km and kcat values for Red1 dependent reduction of (-)-menthone to (+)-neomenthol, were 7.1 μM and 0.60 s-1, respectively. The kinetic parameters for the reverse reaction, oxidation of (+)-neomenthol to (-)-menthone, were Km 305 μM and kcat 0.91 s-1 (table 3, figure 7).

Chapter 3

90

Table 2. Activity of purified Red1 in buffers A-F. Composition of buffers: A – 40 mM potassium phosphate buffer, 0.1% -mercaptoethanol, pH 7.5, B – 40 mM potassium phosphate buffer, 1 mM ascorbic acid, pH 7.5, C – 40 mM potassium phosphate buffer pH 7.5, D – 20 mM potassium phosphate buffer pH 7.5, E – 40 mM potassium phosphate buffer, 100 mM KCl, pH 7.5, F – 20 mM potassium phosphate buffer, 100 mM KCl. Samples were performed in duplicates.

Buffer Ratio (-)-menthone/(+)-neomenthol Main differences in buffer composition A 4.9 Reducing agent -mercaptoethanol B 6.1 Reducing agent ascorbic acid C 4.3 High buffer salt concentration D 4.3 Low buffer salt concentration E 3.9 High buffer salt concentration and added KCl F 3.3 Low buffer salt concentration and added KCl

Table 3. Kinetic parameters for conversion of (-)-menthone and (+)-neomenthol by Red1.

Substrate Km(µM)

kcat(s-1)

kcat/Km(M-1s-1)

(-)-menthone 7.1 ± 1.4 0.60 83620 (+)-neomenthol 305 ± 100 0.91 2990 Menthone:neomenthol reductase from Mentha x piperitea (Davis, Ringer et al. 2005)

674 0.06 89

0 10 20 30 40 500

20

40

60

80

100 300 500 700 900

(-)-Menthone(+)-Neomenthol

M Substrate

pmol

(-)-

conv

ersi

on p

rodu

ct/s

Figure 7. Michaelis-Menten curves for Red1 dependent conversion of (-)-menthone ( ) and (+)-neomenthol ( ).


91

Concluding remarks

Studies of the biosynthesis of natural compounds in Artemisia annuahas so far focused mostly on the biosynthesis of the antimalarial drug artemisinin. Previous studies involving Red1 pointed it out as an efficient and likely reducer of the ketone in the sesquiterpenoid artemisinin intermediate dihydroartemisinic aldehyde (Rydén et al., 2010). However, Artemisia annua, a member of the Asteraceaefamily well known for its sesquiterpenoid biosynthesis, also contains a range of other valuable and interesting terpenoids (table 1). The potential of Red1 to convert important monoterpenoids and sesquiterpenoids, was investigated while keeping in mind the natural metabolic profile of A. annua (table 1) (Bhakuni et al., 2001). To investigate a possible biotechnological use of Red1, monoterpenoids as well as sesquiterpenoids other than those found in A. annua, were selected and tested for suitability as substrates (table 1). Of all the tested monoterpenoid substrates, (-)-menthone proved to be the most efficiently converted. Hence, further kinetic characterization focused on (-)-menthone as substrate. Red1 shares sequence homology with the short chain dehydrogenase/reductase (SDR) protein family (Kallberg et al., 2002), and in particular reductases from mint (Davis et al., 2005) (figure 2). Common to SDRs is their action on molecules such as monoterpenoids (C10 compounds) rather than the sesquiterpeneoids (C15 compounds) (Kallberg et al., 2002). Conversion assays confirmed the ability of Red1 to reduce and oxidize short chain molecules. From the results summarized in table 1, it can be concluded that Red1 has broad substrate specificity (figure 1a).

(-)-Menthol is precent in A. annua in discrete amounts, and due to its industrial importance its precursor, (-)-menthone, was used as a substrate for Red1 (table 1). Crude protein extracts containing Red1 converted (-)-menthone to (-)-menthol (figure 3a, b). However, product specificity was shifted from (-)-menthol to (+)-neomenthol, when purified protein was used (figure 3c), indicating the

Chapter 3

92

interference of E.coli proteins in the position of the chemical equilibrium between (-)-menthone and (-)-menthol. Another interesting feature of Red1 is its chemical equilibrium between (-)-menthone, (-)-menthol and (+)-neomenthol, which show a capability of producing (-)-menthol from (+)-neomenthol (figure 5a, b). This makes Red1 an interesting enzyme for biotechnology, as it is desirable to be able to convert (+)-neomenthol to (-)-menthol. The acceptance of (-)-menthol (figure 6a, b) and refusal of (+)-menthol as a substrate in a dehydrogenase reaction, further supports the enantioselective product formation and substrate specificity.

The capability of Red1 to efficiently convert (-)-menthone to (+)-neomenthol (table 3) may also be interesting for plant protection. The intrinsic value of (+)-neomenthol lies in its antimicrobial effect and it has been shown that the compound has a plant protective action (Choi et al., 2008). Albeit (+)-neomenthol has not been detected in A. annua, the usage of Red1 for biotechnological purposes may be warranted as deemed from the kinetic data (table 3). Production of (+)-neomenthol was accomplished in A. thaliana by overexpression of a (-)-menthone reductase from pepper (Capsicum annuum), which led to enhanced protection against Pseudomonas syringae pv tomato DC3000 and Hyaloperonospora parasitica isolate Noco2. Our results show that Red1 has a kcat/Kmvalue that is 939 fold more efficient than that of (-)-menthone: (+)-neomenthol conversion by the corresponding M. piperitea enzyme (Davis et al., 2005), a plant species specialized in the biosynthesis of menthone and menthol related monoterpenoids. The oxidation of (+)-neomenthol to (-)-menthone by Red1 is unlikely to occur at any significant level, as the kcat/Km value of the reducing reaction is 28 fold higher than the kcat/Km value for the oxidative reaction (table 3). Furthermore, the Km value of the oxidative reaction is in the unphysiological range of 305 μM. Introduction of Red1 in planta,could hence contribute significantly to enhanced resistance against pathogens due to its catalytic efficiency, provided there is (-)-menthone available in the host.


93

Materials and Methods

Chemicals

(+)-Borneol, (-)-borneol, (-)-camphor, (+)-camphor, (-)-carvone , -humulene, neomenthol and (-)-limonene were purchased from Extrasynthèse, Genay, France. (+)-Menthol was purchased from Tokyo Chemical Industry, (-)-menthol from Fluka, (-)-menthone and (+)-nootkatone were purchased from Fluka. (+)-Dihydrocarvone, germacrone, 1- -terpineol and -thujone were synthesized by the Department of Pharmaceutical Biology, Groningen, the Netherlands. A mixture of -thujone and ß-thujone was purchased from Roth. (+)-Carvone, citral 95% cis-trans mixture were purchased from Sigma-Aldrich. Isopiperitenone was obtained as described by Lucker et al. (2004).

GC-MS

Analyses were performed on a HP5890 series II gas chromatograph coupled to an HP5972A mass selective detector (70eV) equipped with a ZB-5MS column (30 m x 0.25 mm i.d., film thickness 0.25 μm, Phenomenex, ordered at Bester, Amstelveen, The Netherlands). Inlet temperature was set to 250 °C and the detection temperature to 290 °C, with a constant helium flow of 1.0 ml/min. Samples were analyzed with the temperature program 55 °C for 4 min, followed by a ramp of 5 °C/min to 220 °C, followed by a ramp of 20 °C to 290 °C with a hold for 4.5 min. Peaks were identified using Wiley mass spectral library, NIST library, internal libraries or via authentic standards.

Sequencing

Constructs were sequenced by the commercial service of Macrogen using BigDye terminator cycling conditions.

Chapter 3

94

Kinetic analysis

The kinetic parameters Km and Vmax were calculated with GraphPad Prism 5 for Windows (GraphPad Software Inc.), using standard settings for non-linear regression curve fitting in Michaelis-Menten mode.

Plant material

Plants were grown for collection of flowers using seeds (UG2006) obtained from University of Groningen (Wallaart et al., 1999). Seeds were sown the 3rd of August 2006 and grown under green house conditions using potting compost. Plants were watered as necessary without addition of fertilizer. When the plants started to flower displaying a range from buds to late flowering with pollen, the flowers were collected in four categories sorted after bud (F1), young flower (F2), old flower (F3) and old flower with pollen (F4). Flowers were frozen in liquid nitrogen and stored at -80 °C. Plant leaf material was picked the 28th of April 2003 and stored at -80 °C (Bertea et al., 2006).

Construction of cDNA and cloning of full length gene red1

A glandular trichome specific cDNA library previously constructed was further sequenced (Bertea et al., 2006). Obtained sequences were compared with the literature, using batch BlastX with standard settings. The EST fragments were additionally analyzed by alignment with EST libraries from tomato, potato, rice, A. thaliana,lettuce and sunflower. It was argued that fragments having high similarity with EST sequences from lettuce and sunflower, which both belong to the family Asteraceae, but comparatively low similarity with the non-Asteraceae family members potato, tomato and rice, would be specific for Asteraceae species and likely to be involved in terpenoid biosynthesis. Based on the combined information from the two alignment approaches, one gene fragment


95

named red1 was selected for further analysis and 5’ extension. Approximately 100 mg frozen flowers (stage F4) and mature leaves (O leaf) were ground in liquid nitrogen. Total RNA was isolated using TriPure (Roche Applied Science) and 1 μg of the isolated F4 and O leaf total RNA were used for 5’ RACE cDNA amplification, using the Smart RACE cDNA amplification kit from Clontech. PCR was performed on an aliquot of the obtained F4 cDNA, using the internal primer Red1-RACE1 5’ cgatttctctcattcgtgaggtgttcaatatcttggagc 3’ and the universal primer mix from the Smart RACE cDNA amplification kit. The PCR product was purified from an agarose gel and ligated into the sequencing vector pGEMT-Easy (Promega) and sequenced. The full length gene was isolated form O leaf cDNA with the primers FL5-AseI-red1 5’ accttggattaatatgtcatatgcaaccgagaaagatg 3’ and FL3-red1-XhoI 5’ accttggctcgagtcaaaatgaggttatttgcattcgac 3’, using Phusion DNA polymerase (Finnzymes) under reaction conditions recommended by the manufacturer (Finnzymes) and 1 μl cDNA. The 953 bp PCR product was purified using the QIAquick PCR Purification Kit of Qiagen, and cloned into pGEMT for propagation and sequencing. Confirmed inserts were excised using AseI and XhoI and ligated into the expression vectors pET26b+ and pET28a+, previously restricted with NdeI and XhoI, forming the constructs pET26b+-red1 and pET28a+red1. Constructs were cloned into E. coli DH5 for propagation, where after plasmids were isolated and transformed into the expression host E. coli BL21 (DE3)-Rosetta2 following manufacturer’s instructions (Novagen). The nucleotide sequence for red1 has been deposited in the GenBank database under GenBankAccession Number GU167953.

Sequence analysis

The sequence of red1 was analyzed via tBlastX using standard settings (Altschul et al., 1990). The amino acid sequences of selected top hits with confirmed enzymatic function, were extracted and aligned using ClustalW2 (Larkin et al., 2007). Sequences were

Chapter 3

96

aligned using bl2seq for proteins with default settings to obtain identity percentages (Altschul et al., 1990).

Protein expression and purification

Seed cultures with 5 ml LB medium supplemented with 30 μg/ml kanamycine and 34 μg/ml chloramphenicol were initiated using one colony from fresh streak outs, and incubated at 37°C over night with shaking. Main cultures of 200 ml 2xTY (2.4% tryptone, 1.5% yeast extract, 0.75% NaCl) supplemented with antibiotics, as described above, were initiated in 1000 ml Erlenmayer-flasks at a calculated OD600 of 0.02 using aliquots from the seed cultures. Cultures were incubated at 37°C with shaking for 4 hours, and thereafter induced with 200 μl 1M IPTG, followed by incubation at 16°C with shaking for 14 hours. Cytosolic proteins were extracted using Bugbuster (Novagen) following manufacturer’s instructions. Proteins emerging from the pET28a-red1 construct were purified on a nickel column using Ni-NTA resin (Qiagen), and dialyzed at 4°C against buffer G (20 mM KH2PO4, 150 mM KCl and 20% glycerol) using a 14 kDa filter (Viskase Companies, Inc.). Proteins were stored at -20°C. Purity was verified using SDS-PAGE.

Conversion assays

Initial conversion assays to determine substrate specificity were performed in 2 ml aliquots containing 20 μl purified protein or crude protein extract, 40 mM potassium phosphate buffer (pH 7.0), 1 mM DTT, 500 μM NADPH for reductase assays, or NADP+ for dehydrogenase assays and 100 μM substrate. Mixtures were incubated at 30°C with gentle shaking over night. All reactions were performed in duplicates. Protein extracts from E. coli harboring the empty expression vectors pET26b+ and pET28a+ served as negative controls. Conversion assays to determine kinetic parameters contained 1.5 μg to 4.5 μg purified protein, 500 μM NADP+ or NADPH, 5-50 μM (-)-menthone or 50-1000 μM (+)-neomenthol in


97

buffer F (20 mM KH2PO4, 150 mM KCl, pH 7.0), amounting to a total reaction volume of 900 μl. Samples were incubated for 5 min. at 30°C and reactions were initiated by adding purified Red1. Kinetic assays were incubated for 10 min. at 30°C. The calculated molecular weight of 34,370 Da for Red1 was used to determine the molar amount of protein added in the assays. Enzyme reactions were stopped on ice and 45 nmol cis-nerolidol was added as an internal standard to samples. Samples were vortexed and 2 ml ethyl acetate was added, followed by vigorous mixing. Samples were centrifuged until complete phase separation was achieved. The organic phase was collected and dried with anhydrous Na2SO4 and analyzed by GC-MS as described above. Kinetic assay samples were prepared by adding 12 nmol cis-nerolidol as internal standard, extracting twice with 600 μl ethyl acetate, and concentrating samples under a gentle nitrogen flow. Samples were analyzed by GC-MS, as described above. Standard curves for quantitative analysis were constructed from authentic standards using (+)-neomenthol and (-)-menthone.

Chapter 3

98

References

Ahmad, A., & Misra, L. N. (1994). Terpenoids from Artemisia annua and constituents of its essential oil. [doi: DOI: 10.1016/0031-9422(94)85021-6]. Phytochemistry, 37(1),183-186.

Altschul, S. F., Gish, W., Miller, W., Myers, E. W., & Lipman, D. J. (1990). Basic local alignment search tool. J Mol Biol, 215(3), 403-410.

Banthorpe, D. V., Baxendale, D., Gatford, C., & Williams, S. R. (1971). Monoterpenes of some Artemisia and Tanacetum species grown in England. Planta Med, 20(2), 147-152.

Banthorpe, D. V., Charlwood, B. V., Greaves, G. M., & Voller, C. M. (1977). Role of geraniol and nerol in the biosynthesis of Artemisia ketone. [doi: DOI: 10.1016/S0031-9422(00)88788-9]. Phytochemistry, 16(9), 1387-1392.

Bertea, C. M., Voster, A., Verstappen, F. W., Maffei, M., Beekwilder, J., & Bouwmeester, H. J. (2006). Isoprenoid biosynthesis in Artemisia annua: cloning and heterologous expression of a germacrene A synthase from a glandular trichome cDNA library. Arch Biochem Biophys, 448(1-2), 3-12.

Bhakuni, R. S., Jain, D. C., Sharma, R. P., & Kumar, S. (2001). Secondary metabolites of. Artemisia annua and their biological activity. Current Science (Bangalore), 80(1), 35-48.


Chalchat, J. C., Garry, R. P., Michet, A., Gorunovic, M., & Stosic, D. (1991). A Contribution to Chemotaxonomy of Artemisia-Annua L. Asteraceae. Acta Pharmaceutica Jugoslavica, 41,233-236.


99

Choi, H. W., Lee, B. G., Kim, N. H., Park, Y., Lim, C. W., Song, H. K., et al. (2008). A role for a menthone reductase in resistance against microbial pathogens in plants. Plant Physiol, 148(1), 383-401.

Covello, P. S., Teoh, K. H., Polichuk, D. R., Reed, D. W., & Nowak, G. (2007). Functional genomics and the biosynthesis of artemisinin. Phytochemistry, 68(14), 1864-1871.

Davis, E. M., Ringer, K. L., McConkey, M. E., & Croteau, R. (2005). Monoterpene metabolism. Cloning, expression, and characterization of menthone reductases from peppermint. Plant Physiol, 137(3), 873-881.

Geissler, R., Brandt, W., & Ziegler, J. (2007). Molecular modeling and site-directed mutagenesis reveal the benzylisoquinoline binding site of the short-chain dehydrogenase/reductase salutaridine reductase. Plant Physiol, 143(4), 1493-1503.

Goel, D., Singh, V., Ali, M., Mallavarupu, G., & Kumar, S. (2007). Essential oils of petal, leaf and stem of the antimalarial plant Artemisia annua. [10.1007/s11418-006-0112-9]. Journal of Natural Medicines, 61(2), 187-191.

Kallberg, Y., Oppermann, U., Jornvall, H., & Persson, B. (2002). Short-chain dehydrogenases/reductases (SDRs). Eur J Biochem, 269(18), 4409-4417.

Larkin, M. A., Blackshields, G., Brown, N. P., Chenna, R., McGettigan, P. A., McWilliam, H., et al. (2007). Clustal W and Clustal X version 2.0. Bioinformatics, 23(21), 2947-2948.

Lawrence, B. M. (1982). Annual wormwood oil. Progr. Essential Oil, 7, 43-44.


Lucker, J., Schwab, W., Franssen, M. C., Van Der Plas, L. H., Bouwmeester, H. J., & Verhoeven, H. A. (2004). Metabolic engineering of monoterpene biosynthesis: two-step

Chapter 3

100

production of (+)-trans-isopiperitenol by tobacco. Plant J, 39(1), 135-145.


Persson, B., Kallberg, Y., Oppermann, U., & Jornvall, H. (2003). Coenzyme-based functional assignments of short-chain dehydrogenases/reductases (SDRs). Chem Biol Interact, 143-144, 271-278.

Ringer, K. L., McConkey, M. E., Davis, E. M., Rushing, G. W., & Croteau, R. (2003). Monoterpene double-bond reductases of the (-)-menthol biosynthetic pathway: isolation and characterization of cDNAs encoding (-)-isopiperitenone reductase and (+)-pulegone reductase of peppermint. Arch Biochem Biophys, 418(1), 80-92.


Rydén, A.-M., Ruyter-Spira, C., Quax, W., Osada, H., Muranaka, T., Kayser, O., et al. (2010). The molecular cloning of dihydroartemisinic aldehyde reductase and its implication in artemisinin biosynthess in Artemisia annua. Planta Medica, In press.


Wallaart, T. E., van Uden, W., Lubberink, H. G., Woerdenbag, H. J., Pras, N., & Quax, W. J. (1999). Isolation and identification of dihydroartemisinic acid from artemisia annua and its


101

possible role in the biosynthesis of artemisinin. J Nat Prod, 62(3), 430-433.

Woerdenbag, H. J., Bos, R., Salomons, M. C., Hendriks, H., Pras, N., & Malingré, T. M. (1993). Volatile constituents of Artemisia annua L. (Asteraceae). Flavour and Fragrance Journal, 8(3), 131-137.

Woerdenbag, H. J., Pras, N., Chan, N. G., Bang, B. T., Bos, R., van Uden, W., et al. (1994). Artemisinin, Related Sesquiterpenes, and Essential Oil in Artemisia annua During a Vegetation Period in Vietnam. Planta Med, 60(3), 272-275.


Chapter 3

102

Chapter 4

The molecular cloning of dihydroartemisinic aldehyde

reductase and its implication in artemisinin biosynthesis in

Artemisia annua

Anna-Margareta Rydén1,2,3, Carolien Ruyter-Spira2, Wim Quax1,Hiroyuki Osada4, Toshiya Muranaka3, Oliver Kayser5 and Harro Bouwmeester6

1 Pharmaceutical biology, University of Groningen, Groningen, The Netherlands2 Plant Research International, Wageningen, The Netherlands 3 Kihara Institute for Biological Research, Yokohama City University, Yokohama, Japan 4 Chemical Biology Department, Antibiotics Laboratory, Advanced Science Institute, RIKEN, Saitama, Japan 5 Technical Biochemistry, Technical University of Dortmund, Dortmund, Germany 6 Laboratory of Plant Physiology, Wageningen University, Wageningen, The Netherlands

Adapted from Planta Medica. In press (2010).

Chapter 4

104

Abstract

A key point in the biosynthesis of the antimalarial drug artemisinin is the formation of dihydroartemisinic aldehyde which reprecents the key difference between chemotype specific pathways. This key intermediate is the substrate for several competing enzymes some of which increase the metabolic flux towards artemisinin and some of which – as we show in the precent study – may have a negative impact on artemisinin production. In an effort to understand the biosynthetic network of artemisinin biosynthesis, extracts of Artemisia annua (A. annua) flowers were investigated and found to contain an enzyme activity competing in a negative sense with artemisinin biosynthesis. The enzyme, Red1, is a broad substrate oxidoreductase belonging to the short chain dehydrogenase/reductase family with high affinity for dihydroartemisinic aldehyde and valuable monoterpenoids. Spatial and temporal analysis of cDNA revealed red1 to be trichome specific. The relevance of Red1 to artemisinin biosynthesis is discussed.

Dihydroartemisinic aldehyde reductase

105

Introduction

Since the discovery four decades ago that artemisinin from the plant Artemisia annua L. (A. annua) is the active compound against malaria there has been a keen interest in synthesizing it or derivatives thereof (Rydén et al., 2007). More recently biochemical studies revealed the probable intermediates in the biosynthetic pathway (Bertea et al., 2005; Wallaart et al., 1999; Wallaart et al., 1999). Given the leads from these biochemical studies, EST libraries were sequenced in the quest for genes involved in the pathway (Mercke et al., 2000; Ro et al., 2006; Teoh et al., 2006; Wallaart et al., 2001) complemented with a reverse genetics approach (Zhang et al., 2008). Currently, breeding programmes as well as microbial heterologous pathway expression studies are developed to satisfy the need for a stable supply of the drug. The focus on artemisinin derived from the plant has been on growth condition improvement, harvesting methods and marker assisted breeding. Microbial production systems on the other hand have focused on identification of the biosynthetic genes in the linear pathway. These efforts led to the identification of the terpene cyclase amorpha-4,11-diene synthase which catalyses the first committed step of the biosynthetic pathway of artemisinin (figure. 1) (Bouwmeester et al., 1999; Wallaart et al., 2001). The second step is catalysed by the cytochrome P450 Cyp71av1 that oxidizes amorpha-4,11-diene at C12 in three consecutive steps (figure. 1) (Ro et al., 2006; Teoh et al., 2006). Release of artemisinic aldehyde, which is the product after the second oxygenation step, is allowing Dbr2, a carbon double bond reductase, to reduce the

11(13) carbon double bond thereby producing dihydroartemisinic aldehyde (Zhang et al., 2008) (figure. 1). Thus a solid understanding of the production of the early and late intermediates in the pathway has been developed. However, given the global aim of increased production of artemisinin, it is important to study bottlenecks in the pathway both in planta and in heterologous systems as well as the impact of competing biosynthetic pathways on artemisinin production.

Chapter 4

106

OP

OP

O O

OO

O

Amds

H

H

farnesyl diphosphate

Cyp71AV1

H

H

OH

Cyp71AV1

H

H

Oamorpha-4,11-diene artemisinic alcohol artemisinic aldehyde

H

H

O

OH

Cyp71AV1

H

H

O

Red1

H

H

OHdihydroartemisinic acid dihydroartemisinic

aldehydedihydroartemisinicalcohol

Dbr2

artemisinin

Figure 1. Biosynthetic pathway of artemisinin and the activity of Red1.

As part of an effort to understand the isoprenoid metabolism in A. annua, an EST library was employed to isolate genes involved in the biosynthetic pathway (Bertea et al., 2006). From the EST library, an oxidoreductase named red1 was identified and cloned. Here we report on the kinetic characteristics of Red1 and its role in the biosynthetic pathway of artemisinin. The impact of Red1 on this biosynthetic network is discussed.


Several semi-quantitative PCR reactions were performed to compare the presence of transcripts in different tissues of A. annua (figure 2). The transcription patterns of other genes than red1 relevant to the biosynthetic pathway of artemisinin were also investigated to be able to compare trends in transcription. Along with a spatial distribution of the transcripts, the temporal variable was investigated dividing tissues into two categories for leaves: young leaf and old leaf and


107

amds

cyp71av1

fds

actin1

red1

Old leaf

Young leaf

Stem

Flower

Flower 4

Flower 3

Flower 2

Flower 1

Figure 2. Semi-quantitative PCR on cDNA from Artemisia annua. red1– reductase 1, amds – amorpha-4,11-diene synthase, cyp71av1– cytochrome P450 71av1, fds – farnesyl diphosphate synthase.

four categories for flowers. In all tissues, fds was transcribed at similar levels (figure 2). This is an expected result as the product farnesyl diphosphate formed by the enzyme encoded by fds, is a common precursor of many ubiquitous molecules such as sterols and ubiquinones (Aharoni et al., 2005). The transcription of amds, which catalyses the first committed step of the artemisinin biosynthetic pathway by creating the amorphane skeleton from farnesyl diphosphate (Bouwmeester et al., 1999; Mercke et al., 2000; Wallaart et al., 2001), displayed a tissue and time dependent profile (Kim et al., 2008). The transcript levels of amds gradually decreased to undetectable levels in time in flowers and leaves while transcription in stems was not observed. The transcription of cyp71av1 did not follow the same strict transcription profile as amdshowever it was confined in the same manner as amds to flowers and young leaves. In contrast to amds, cyp71av1 was transcribed at a stable level throughout the life span of the flower. Transcription of red1 occurred in flowers in a time dependent manner: it was high in buds, young flowers and old flowers with pollen but absent in completely open flowers without pollen (figure 2). Just as for transcription of amds and cyp71av1, there was no red1 transcription in the stem. Just as for amds, red1 transcription was high in young

Chapter 4

108

leaves but in contrast to amds, it was also high in old leaves (figure 2). This is an expected result as it has been shown that amorpha-4,11-diene decease in time with maturing plant while dihydroartemisinic acid and artemisinin levels increase (Kim et al., 2008; Wallaart et al., 2000).

Analysis of red1 with BlastX produced the enzymatically confirmed hits Mentha x piperita (M. piperita) (-)-isopiperitenone reductase, M.piperita (-)-menthone:(+)-neomenthol reductase (MNR) (Davis et al., 2005), M. piperita menthol dehydrogenase (MMR) (Davis, et al., 2005) and Papaver bracteatum salutaridine reductase (Geissler et al., 2007). These enzymes are categorized as belonging to the short chain dehydrogenase reductase (SDR) family (Kallberg et al., 2002; Oppermann et al., 2003; Persson et al., 2009). The common features of the SDR family are the N-terminal cofactor binding motif GxxxGxG (motif 1) and the catalytic domain YxxxK (motif 2) as well as the preference of NADP(H) over NAD(H) due to their basic lysine residue in motif 1 (Persson et al., 2003) which categorizes them as classical SDR enzymes. Consistent with the BlastX hits and biochemical studies, Red1 is categorized as belonging to the SDR family. Currently three cytosolic proteins involved in the biosynthetic pathway of artemisinin have been reported: Amds (Bouwmeester et al., 1999; Mercke et al., 2000; Wallaart et al., 2001), alcohol dehydrogenase 1 (Aldh1) (Teoh et al., 2007) and double bond reductase 2 (Dbr2) (Teoh et al., 2007; Zhang et al., 2008).

Biochemical studies of Red1 revealed it to be a broad substrate SDR family protein accepting among others menthone, menthol and neomenthol as substrates (Rydén et al., 2010) (table 1). Red1 accepts dihydroartemisninic aldehyde as a substrate converting it to dihydroartemisinic alcohol, but the closely similar substrate artemisinic aldehyde is not converted despite its close similarity with and greater chemical reactivity than dihydroartemisinic aldehyde (table 1, figure 3c-d).


109

Table 1. Apparent kinetic parameters for cytosolic reductases from A. annua, Mentha x piperita and Carpoglyphus lactis.Aldh1 – alcohol dehydrogenase 1 (A. annua) (Teoh et al., 2007), Dbr2 – Double bond reductase 2 (A. annua) (Zhang et al., 2008), GeDH – Geraniol dehydrogenase (C. lactis)(Noge, 2008), MNR – Menthone:neomenthol reductase (Mentha x piperita) (Davis et al., 2005), MMR – Menthone:menthol reductase (Mentha x piperita) (Davis et al., 2005), Red1 – Dihydroartemisinic aldehyde reductase (A. annua) (this study), n. r. – not reported.

Protein vmax(pmol,s-1)

Km(μM)

kcat(s-1)

kcat/Km(M-1, s-1) Substrate Product

Aldh1 n. r. 2.58 1.53 593000 AAA AA n. r. 8.79 7.74 881000 DHAAA DHAA

Dbr2 n. r. 19 2.6 0.14 AAA DHAAA

n. r. 790 1.8 0.0023 2-Cyclohexen- 1-one Cyclohexanone

n. r. 650 0.86 0.0013 (+)-Carvone (-)-Isodihydro-carvone,

(-)-dihydrocarvone

GeDH n. r. 51.0 2996 58800000 Geraniol Geranial

MNR n. r. 674 ± 78.0 0.06 89 (-)-Menthone (+)-Neomenthol n. r. > 1000 n. r. n. r. (+)-Isomenthone (+)-Isomenthol

n. r. > 1000 n. r. n. r. (+)-Neomenthol (-)-Menthone

MMR n. r. 3.0 ± 0.6 0.6 200000 (-)-Menthone (-)-Menthol n. r. 41 ± 5.0 n. r. n. r. (+)-Isomenthone (+)-Neoisomenthol

Red1 24 ± 2.8 67 ± 15 0.28 4119 DHAAA DHAAOH n. r. > 1000 n. r. n. r. DHAAOH DHAA n. r. > 1000 n. r. n. r. AAA AAOH 52 ± 2.7 7.1 ± 1.4 0.6 83617 (-)-Menthone (+)-Neomenthol

79 ± 10 305 ± 101 0.9 2993 (+)-Neomenthol (-)-Menthone

To further investigate the capability of Red1 to convert aldehyde substrates, perilla aldehyde was used as a substrate. Figure 3a and b

Chapter 4

110

show that Red1 efficiently converts perilla aldehyde into its corresponding alcohol.

a

17.80 18.00 18.20 18.40 18.60

O

OH

27.60 27.80 28.00 28.20 28.40 28.60 28.80 29.00 29.20

c

OH

H

H

17.80 18.00 18.20 18.40 18.60

b O

0 27.40 27.60 27.80 28.00 28.20 28.40 28.60 28.80 29.00 29.20

d

O

H

H

Figure 3. Substrate and product profile of Red1. a) Purified his-tagged Red1 with perilla aldehyde as substrate, b) Empty vector control with perilla aldehyde as substrate, c) Purified his-tagged Red1 with dihydroartemisinic aldehyde as substrate, d) Empty vector control with dihydroartemisinic aldehyde as substrate.


111

Also dihydroartemisinic alcohol, artemisinic aldehyde and artemisinic alcohol were investigated for their suitability as substrates for Red1 (data not shown), but no conversion activity could be detected using these substrates.

The apparent kinetic parameters of Red1 show that the monoterpenoid (-)-menthone is a preferred substrate (table 1) with Km, kcat and kcat/Km values of 7.1 μM, 0.6 s-1 and 83617 M-1, s-1,respectively. The sesquiterpenoid dihydroartemisinic aldehyde on the other hand is, albeit less efficiently than (-)-menthone, relatively efficiently converted into dihydroartemisinic alcohol. Interestingly the reverse reaction, oxidation of dihydroartemisinic alcohol to dihydroartemisinic aldehyde, could not be detected (table 1). The apparent kinetic parameters of Red1 converting dihydroartemisinic aldehyde were calculated to be Km 67μM, kcat 0.28 s-1 and kcat/ Km4119 M-1, s-1. These data show that Red1 is capable of efficient and specific reduction of ketone/aldehyde groups in dihydroartemisinic aldehyde as well as monoterpenoids.

Concluding remarks

The biosynthesis of artemisinin in A. annua is intensively studied due to its activity against the malaria parasite (Covello et al., 2007; Rydén et al., 2007; Zeng et al., 2008). The first committed step in the biosynthetic pathway, the cyclisation of farnesyl diphosphate to amorpha-4,11-diene (Bouwmeester et al., 1999; Mercke et al., 2000; Picaud et al., 2006; Wallaart et al., 2001) and the subsequent oxygenation to alcohol, aldehyde and acid are well characterized through biochemical studies on the plant (Bertea et al., 2005), cloning of responsible genes (Bouwmeester et al., 1999; Mercke et al., 2000; Teoh et al., 2006; Wallaart et al., 2001; Zhang et al., 2008) and heterologous characterisation of pathway intermediates (Ro et al., 2006; Zhang et al., 2008) (figure 1). There are two chemotypes of A. annua one the artemisinic acid the other a dihydroartemisinic acid accumulating chemotype, which have been shown to have low and

Chapter 4

112

high artemisinin production, respectively (Wallaart et al., 2000). The cloning of dbr2 and its efficient conversion of artemisinic aldehyde to dihydroartemisinic aldehyde (figure 1) suggests that the main chemotype shift occurs at the aldehyde level rather than at the alcohol or acid levels (Zhang et al., 2008) as has been postulated (Bertea et al., 2005). Expression of Dbr2 and other proteins necessary to reconstitute the artemisinin biosynthetic pathway in yeast produced a strain yielding 11.8 μg/mL artemisinic acid and 15.7 μg/mL dihydroartemisinic acid, as compared to the control strain lacking Dbr2 expression which produced 29.4 μg/mL artemisinic acid (Zhang et al., 2008). As the biosynthetic pathway has been removed from its context while reconstituted in the heterologous host, it is difficult to establish whether the comparatively high production of artemisinic acid even in strains expressing Dbr2 is due to imbalance in expression levels compared to the situation in the plant and/or due to an incomplete palette of expressed biosynthetic genes. Even so, the suitability of yeast as a factory for production of artemisinic acid has been previously established (Ro et al., 2008; Ro et al., 2006). The fate of the late intermediates artemisinic acid and dihydroartemisinic acid is less clear. There is evidence that points to a non-enzymatic conversion of dihydroartemisinic acid to artemisinin in planta (Brown et al., 2003; Brown et al., 2004).

The expression of Dbr2 in the yeast strain is interesting evidence for the chemotypic shift although it is to early to say that it is the only point at which the 11(13) carbon double bond reduction of amorpha-4,11-diene occurs. Precently there are no data for the kinetic parameters of Dbr2 catalysed conversion of dihydroartemisinic aldehyde to artemisinic aldehyde. It is possible that additional enzymes other than Cyp71av1 and to a minor extent Aldh1 exist, that convert dihydroartemisinic aldehyde to dihydroartemisinic acid. It is thus difficult to estimate the impact of Red1 in the biosynthetic network of artemisinin. Currently a system for stable transformation of A. annua is being developed to be able to


113

address this question. With the cloning of red1, there are a total of four known enzymes competing for the substrate dihydroartemisinic aldehyde, the other three being the alcohol dehydrogenase Aldh1 (Teoh et al., 2007), the cytochrome P450 Cyp71av1 (Rydén et al., 2009; Teoh et al., 2006) and Dbr2 (Zhang et al., 2008). Only for Aldh1 have the apparent kinetic data been reported (table 1). These data suggest that under those experimental conditions and employing NADP+ as cofactor, Aldh1 is more efficiently converting dihydroartemisinic aldehyde into dihydroartemisinic acid than Red1 is converting dihydroartemisinic aldehyde into dihydroartemisinic alcohol. Earlier studies performed on cytosolic fractions from A. annua report that dihydroartemisinic aldehyde was reduced to dihydroartemisinic alcohol as well as oxidized to dihydroartemisinic acid (Bertea et al., 2005). Production of dihydroartemisinic alcohol can be seen as detrimental for the biosynthetic pathway of artemisinin as this metabolite is a less favored substrate of the major oxidizing enzyme Cyp71av1 than artemisinic aldehyde (Rydén et al., 2009). Also the chemical properties of dihydroartemisinic alcohol versus dihydroartemisinic aldehyde make the alcohol a less suitable substrate as it is less reactive than the aldehyde. The higher activation energy barrier of dihydroartemisinic alcohol compared to the dihydroartemisinic aldehyde is one explanation for the refusal of the metabolite as a substrate by Red1. Production of dihydroartemisinic alcohol will in a global sense lower the rate of production of dihydroartemisinic acid. Although there might be other enzymes involved in the reduction and oxidation of dihydroartemisinic aldehyde in planta, the overall scheme of a major flow from artemisinic aldehyde via dihydroartemisinic aldehyde to the chemically more stable dihydroartemisinic alcohol was established. The cloning and functional characterization of the broad substrate oxidoreductse red1 is a reminder that to optimize a biosynthetic pathway in the original host, it is necessary to consider not only the forward biosynthetic route but also competing metabolic pathways. The ability of Red1 to convert dihydroartemisinic aldehyde to dihydroartemisinic alcohol makes it an interesting target

Chapter 4

114

for knock down studies. Such investigation should clarify whether the function of dihydroartemisinic alcohol, given that it a less preferred substrate than artemisinic alcohol and dihydroartemisnic aldehyde for Cyp71av1, is as a storage molecule to prevent accumulation of the more reactive and potentially dangerous dihydroartemisinic aldehyde. The removal of Red1 from the biosynthetic network is also anticipated to assist in discovery of other genes that encode proteins acting on the biosynthetic intermediates of artemisinin. If Red1 has a role as a monitor protein keeping dihydroartemisininc aldehyde at suitable, non-toxic levels, it is likely that the removal of Red1 from the network will upregulate or downregulate other genes in a stress response against accumulated dihydroartemisinic aldehyde. However, in the long run it is anticipated that silencing of Red1 may lead to increased production of artemisinin in A. annua.

Material and Methods

Plant materials

Plants, originally identified by Wallaart et al. (1999), were grown for collection of flowers, leaf and root material using seeds obtained from University of Groningen. Seeds are deposited in the herbarium De Kruidhof in Buitenpost, the Netherlands, under registration number GR001. The A. annua line belongs to the high artemisinin chemotype and has the characteristic pattern of a high dihydroartemisinic acid to artemisinic acid ratio. Seeds were sown the 3rd of August 2006 and seedlings grown under green house conditions at 21/18°C (16/8 h) using potting compost. Daylight was supplemented with artificial light (SON-T AGRO) during the high-temperature period. Plants were watered as necessary without addition of fertilizer. When the plants started to flower, flowers were collected in four categories: bud (F1), young flower (F2), old flower (F3) and old flower with pollen (F4). Green young leaves and mature leaves (O leaf) were collected along with stem material. Plant


115

material was kept on ice during handling, then frozen in liquid nitrogen and stored at -80 °C.

Chemicals

Neat artemisinic alcohol, artemisinic aldehyde, dihydroartemisinic alcohol and dihydroartemisinic aldehyde were synthesized as described by Bertea et al. (2006). Neat perilla aldehyde and perilla alcohol were commissioned to the Department of Pharmaceutical Biology, Groningen, the Netherlands. Primers for semi-quantitative PCR were custom synthesized by BEX, Japan (table 2).

Table 2. Primers used for semi-quantitative PCR in this study.

Gene Primer pair Product size Cycle number

red1 5’ tac gga act aaa gga gta ac 3’ 5’ tga ggt tat ttg cat tcg ac 3’ 477 bp 35

amds 5’ cac aag gaa gag ctc agc cat gtg tgc a 3’5’ tca ttt agg cgt cga cca agt at acct g 3’ 620 bp 30

cyp71av1 5’ ttg gag cag gca cag aca ctt cct 3’ 5’ acg agt aac aac tca gtc ttt c 3’ 583 bp 35

fds 5’ tat tca ccg ccg aat tgt tc 3’ 5’ aag gat ttc aac acc gct tg 3’ 456 bp 25

actin1 5’ tcc aac tct ggt tct act atc 3’ 5’ ctt cga tct tca tgc tgc tc 3’ 298 bp 30

GC/MS analysis

GC/MS analyses were performed on an Agilent 6890 series gas chromatograph system coupled to a JEOL JMS-SUN 200 mass selective detector in electron impact mode (70 eV) equipped with a guard column (1 m x 0.25 mm i.d., GL-Sciences) and a HP-5MS analytical column (30 m x 0.25 mm i.d., 0.25 μm film). The

Chapter 4

116

temperature program was set to an initial temperature of 60°C followed by a first temperature gradient of 3°C min-1 to 220°C and then a second gradient of 30°C min-1 to 300°C with an endpoint hold of 4 min. Helium flow was set to 0.7 mL min-1, detector temperature and injector temperature to 250°C using 500 scans min-1 with a scan range of 40-350 m/z.

Cloning of red1 and delineation of expression pattern in plant tissues

A glandular trichome specific cDNA library previously constructed was further sequenced (Bertea et al., 2006). Obtained sequences were compared with the literature using batch BlastX with standard settings. The EST fragments were further analyzed by alignment with EST libraries from tomato, potato, rice, Arabidopsis thaliana,lettuce and sunflower. It was argued that fragments having high similarity with EST sequences from lettuce and sunflower which both belong to the family Asteraceae but comparatively low similarity with the non-Asteracea family members potato, tomato and rice would be specific for Asteraceae species and likely to be involved in artemisinin biosynthesis. Based on the combined information from the two alignment approaches, one gene fragment named red1 was selected for further analysis and 5’ extension. Approximately 100 mg frozen flowers (stage F4) and mature leaves (O leaf) were ground in liquid nitrogen. Total RNA was isolated using TriPure (Roche Applied Science) and 1 ug of the isolated F4 and O leaf total RNA were used for 5’ RACE cDNA amplification using the Smart RACE cDNA amplification kit from Clontech. PCR was performed on an aliquot of the obtained F4 cDNA using the internal primer Red1-RACE1 5’ cgatttctctcattcgtgaggtgttcaatatcttggagc 3’ and the universal primer mix from the Smart RACE cDNA amplification kit. The PCR product was purified from an agarose gel and ligated into the sequencing vector pGEMT-Easy (Promega) and sequenced. The full length gene was isolated form O leaf cDNA with the primers FL5-


117

AseI-red1 5’ accttggattaatatgtcatatgcaaccgagaaagatg 3’ and FL3-red1-XhoI 5’ accttggctcgagtcaaaatgaggttatttgcattcgac 3’ using Phusion DNA polymerase (Finnzymes) under reaction conditions recommended by the manufacturer (Finnzymes) using 1 μL cDNA. The 953 bp PCR product was purified using the QIAquick PCR Purification Kit of Qiagen and cloned into pGEMT for propagation and sequencing. Confirmed inserts were excised using AseI and XhoI and ligated into the expression vectors pET26b+ and pET28a+ previously restricted with NdeI and XhoI forming the constructs pET26b+-red1 and pET28a+red1. Constructs were cloned into Escherichia coli (E. coli) DH5 for propagation where after plasmids were isolated and transformed into the expression host E. coli BL21 (DE3)-Rosetta2 following manufacturer’s instructions (Novagen). The nucleotide sequence for red1has been deposited in the GenBank database under GenBank Accession Number GU167953. Semi-quantitative PCR was performed on obtained cDNA using primers listed in table 1 following the recommended reaction conditions of the Taq DNA polymerase manufacturer (Invitrogen). To ensure equal reprecentation of cDNA, actin was employed as a control.

Reductase assays

Initial conversion assays to determine substrate specificity were performed in 2 mL aliquots containing 20 μL purified protein or crude protein extract, 40 mM potassium phosphate buffer (pH 7.0), 1 mM DTT, 500 μM NADPH for reductase assays or NADP+ for dehydrogenase assays and 10 or 100 μM substrate. Mixtures were incubated at 30°C with gentle shaking over night. All reactions were performed in duplicates. Protein extracts from E. coli harboring the empty expression vectors pET26b+ and pET28a+ served as negative controls.

Chapter 4

118

Characterization of recombinant dihydroartemisinic aldehyde reductase

Seed cultures with 5 mL LB medium supplemented with 30 μg/mL kanamycine and 34 μg/mL chloramphenicol were initiated using one colony from fresh streak outs and incubated at 37°C over night with shaking. Main cultures of 200 mL 2xTY (2.4% tryptone, 1.5% yeast extract, 0.75% NaCl) supplemented with antibiotics as described above were initiated in 1000 mL Erlenmayer-flasks at a calculated OD600 of 0.02 using aliquots from the seed cultures. Cultures were incubated at 37°C with shaking for 4 hours and thereafter induced with 200 μL 1M IPTG followed by incubation at 16°C with shaking for 14 hours. Cytosolic proteins were extracted using Bugbuster (Novagen) following manufacturer’s instructions. Proteins emerging from the pET28a-red1 construct were purified on a nickel column using Ni-NTA resin (Qiagen) and dialyzed at 4°C against buffer A (20 mM KH2PO4, 150 mM KCl and 20% glycerol) using a 14 kDa filter (Viskase Companies, Inc.). Proteins were stored at -20°C. Purity was verified using SDS-PAGE. The optimum pH of Red1 was determined to be 7.0 in an assay that included 20 mM phosphate and 60 mM Tris buffers with 150 mM KCl adjusted to pH 6.0 – 9.0 with 0.5 units increments, 1.5 μg purified protein and 60 μM substrate. For characterization of Red1, 300 μL reaction mixtures containing 1.5 μg protein, 500 μM NADPH or NADP+, 20 mM KH2PO4, 150 mM KCl adjusted to pH 7.0 were routinely used. Reductase assays were performed by adding dihydroartemisinic aldehyde as substrate covering the range 10 μM-200 μM or 60 μM-1mM dihydroartemisinic alcohol. Assays were pre-incubated at 30°C for 5 min and reactions initiated by adding Red1 followed by gentle mixing and incubation for 5 min. Reactions were stopped by adding 400 μL ethyl acetate followed by vigorous mixing and storage on ice for one hour before proceeding with two consecutive extractions using 400 μL ethyl acetate per extraction event. Extractions were pooled and evaporated to dryness on ice under a gentle flow of nitrogen gas. Dried samples were resuspended in 60-240 μL ethyl


119

acetate spiked with 10 μM cis-nerolidol as internal standard. Quantification of reaction products was carried out by GC-MS analysis. Control assays were performed with boiled protein. Reactions were performed in triplicates and the response of the internal standard was used to quantify and control reproducibility of the experiments. Reaction products were quantified by comparing peak areas with external standard curves of dihydroartemisinic aldehyde and dihydroartemisinic alcohol. The reaction conditions were chosen to limit conversion of the substrate to less than 10%. Apparent kinetic parameters were calculated with the software GraphPad Prism 5 using non-linear regression analysis in Michaelis-Menten modus (GraphPad Software Inc. San Diego, CA).

Sequence analysis

The amino acid sequences of Dbr2, Aldh1 and Red1 were compared using ClustalW2 (Larkin et al., 2007). BlastX analysis was performed using standard settings (Altschul et al., 1990).

References

Aharoni, A., Jongsma, M. A., & Bouwmeester, H. J. (2005). Volatile science? Metabolic engineering of terpenoids in plants. [doi: DOI: 10.1016/j.tplants.2005.10.005]. Trends in Plant Science, 10(12), 594-602.

Altschul, S. F., Gish, W., Miller, W., Myers, E. W., & Lipman, D. J. (1990). Basic local alignment search tool. J Mol Biol, 215(3), 403-410.


Bertea, C. M., Voster, A., Verstappen, F. W., Maffei, M., Beekwilder, J., & Bouwmeester, H. J. (2006). Isoprenoid

Chapter 4

120

biosynthesis in Artemisia annua: cloning and heterologous expression of a germacrene A synthase from a glandular trichome cDNA library. Arch Biochem Biophys, 448(1-2), 3-12.


Brown, G. D., Liang, G. Y., & Sy, L. K. (2003). Terpenoids from the seeds of Artemisia annua. Phytochemistry, 64(1), 303-323.

Brown, G. D., & Sy, L. K. (2004). In vivo transformations of dihydroartemisinic acid in Artemisia annua plants. Tetrahedron, 60(5), 1139-1159.

Covello, P. S., Teoh, K. H., Polichuk, D. R., Reed, D. W., & Nowak, G. (2007). Functional genomics and the biosynthesis of artemisinin. Phytochemistry, 68(14), 1864-1871.

Davis, E. M., Ringer, K. L., McConkey, M. E., & Croteau, R. (2005). Monoterpene Metabolism. Cloning, Expression, and Characterization of Menthone Reductases from Peppermint. Plant Physiol., 137(3), 873-881.

Davis, E. M., Ringer, K. L., McConkey, M. E., & Croteau, R. (2005). Monoterpene metabolism. Cloning, expression, and characterization of menthone reductases from peppermint. Plant Physiol, 137(3), 873-881.

Geissler, R., Brandt, W., & Ziegler, J. (2007). Molecular modeling and site-directed mutagenesis reveal the benzylisoquinoline binding site of the short-chain dehydrogenase/reductase salutaridine reductase. Plant Physiol, 143(4), 1493-1503.

Kallberg, Y., Oppermann, U., Jornvall, H., & Persson, B. (2002). Short-chain dehydrogenases/reductases (SDRs). Eur J Biochem, 269(18), 4409-4417.

Kim, S. H., Chang, Y. J., & Kim, S. U. (2008). Tissue specificity and developmental pattern of amorpha-4,11-diene synthase (ADS) proved by ADS promoter-driven GUS expression in


121

the heterologous plant, Arabidopsis thaliana. Planta Med, 74(2), 188-193.

Koji Noge, M. K., Naoki Mori, Michihiko Kataoka, Chihiro Tanaka, Yuji Yamasue, Ritsuo Nishida, Yasumasa Kuwahara,. (2008). Geraniol dehydrogenase, the key enzyme in biosynthesis of the alarm pheromone, from the astigmatid mite Carpoglyphus lactis (Acari: Carpoglyphidae). FEBSJournal, 275(11), 2807-2817.

Larkin, M. A., Blackshields, G., Brown, N. P., Chenna, R., McGettigan, P. A., McWilliam, H., et al. (2007). Clustal W and Clustal X version 2.0. Bioinformatics, 23(21), 2947-2948.


Oppermann, U., Filling, C., Hult, M., Shafqat, N., Wu, X., Lindh, M., et al. (2003). Short-chain dehydrogenases/reductases (SDR): the 2002 update. Chem Biol Interact, 143-144, 247-253.

Persson, B., Kallberg, Y., Bray, J. E., Bruford, E., Dellaporta, S. L., Favia, A. D., et al. (2009). The SDR (short-chain dehydrogenase/reductase and related enzymes) nomenclature initiative. Chem Biol Interact, 178(1-3), 94-98.

Persson, B., Kallberg, Y., Oppermann, U., & Jornvall, H. (2003). Coenzyme-based functional assignments of short-chain dehydrogenases/reductases (SDRs). Chem Biol Interact, 143-144, 271-278.


Ro, D. K., Ouellet, M., Paradise, E. M., Burd, H., Eng, D., Paddon, C. J., et al. (2008). Induction of multiple pleiotropic drug

Chapter 4

122

resistance genes in yeast engineered to produce an increased level of anti-malarial drug precursor, artemisinic acid. BMC Biotechnol, 8, 83.


Rydén, A.-M., & Kayser, O. (2007). Chemistry, Biosynthesis and Biological Activity of Artemisinin and Related Natural Peroxides, Bioactive Heterocycles III (pp. 1-31).

Rydén, A.-M., Ruyter-Spira, C., Litjens, R., Takahashi, S., Quax, W. J., Osada, H., et al. (2010). Molecular cloning and characterization of a broad substrate terpenoid oxidoreductase from Artemisia annua. Plant Cell Physiol. In press.

Rydén, A.-M., Ruyter-Spira, C., Van den End, F., Verstappen, F., Quax, W. J., Osada, H., et al. (2010). Oxygen dependent biosynthesis of artemisinin precursors in yeast. J. Biotechn. Submitted.


Teoh, K. H., Reed, D. W., Polichuk, D. R., & Covello, P. S. (2007). World Intellectual Property Organization.


Wallaart, T. E., Pras, N., Beekman, A. C., & Quax, W. J. (2000). Seasonal variation of artemisinin and its biosynthetic precursors in plants of Artemisia annua of different


123

geographical origin: proof for the existence of chemotypes. Planta Med, 66(1), 57-62.



Zeng, Q., Qiu, F., & Yuan, L. (2008). Production of artemisinin by genetically-modified microbes. Biotechnol Lett, 30(4), 581-592.


Chapter 5

Oxygen dependent biosynthesis of artemisinin precursors

in yeast

Anna-Margareta Rydéna, b, Carolien Ruyter-Spirab,c, Francel Verstappenb,c, Fred van den Endd, Wim J. Quaxa, Harro J. Bouwmeesterb, c, Oliver Kaysera,e

a Department of Pharmaceutical biology, University of Groningen, Antonius Deusinglaan 1, Groningen 9713 AV, the Netherlands b Plant Research International, P.O. Box 14, 6700 AA Wageningen, the Netherlands, c currently: Laboratory of Plant Physiology, Wageningen University, P.O. Box 658, 6700 AR Wageningen, the Netherlands d Department of Bioprocess Engineering, Wageningen University, P.O. Box 8129, 6700 EV Wageningen, the Netherlands e Technical Biochemistry, Technical University of Dortmund, Emil-Figge-Str. 68, 44227 Dortmund

Adapted from Journal of Biotechnology. Submitted (2010)

Chapter 5

126

Abstract

The medicinal plant Artemisia annua L. (A. annua) contains artemisinin, a metabolite very effectively eradicating the malaria parasite in patients suffering from malaria. However, the concentration of this endoperoxide sesquiterpene lactone in planta islow (0,2-0,8% of the dry weight) and hence yield and availability of artemisinin are limited. It is therefore desirable to develop new strategies to cope with the shortage. One approach is to use genetically modified micro-organisms and that produce artemisinin or metabolic precursors of artemisinin that can be chemically converted to artemisinin. In this report, the suitability of Saccharomyces cerevisiae (S. cerevisiae) as a host for the heterologous production of artemisinin precursors was investigated. We focused on the possibilities to use a combination of heterologous, A. annua derived, and endogenous yeast genes and controlled environmental conditions to affect product specificity. Hereto, two key enzymes in the artemisinin pathway, amorpha-4,11-diene synthase and the cytochrome P450 Cyp71av1 were expressed in S.cerevisiae. The production of known intermediates from the artemisinin pathway was confirmed, but under restrictive oxygen conditions a shift in product pattern occurred from artemisinic acid to dihydroartemisinic acid. We postulate that under less optimal conditions for Cyp71av1 catalysis, an endogenous yeast reductase is able to reduce the exocyclic double bond of one of the enzyme intermediates, hence leading to reduced artemisinin precursors. For the first time, we show that Cyp71av1 can also catalyze the oxidation of dihydroartemisinic aldehyde to dihydroartemisinic acid. Dihydroartemisinic acid is a preferred product, as it is one step closer to artemisinin than artemisinic acid, and can be more easily chemically converted to artemisinin.

Fermentation studies

127

Introduction

Artemisia annua L. (A. annua) is an annual herb that produces the antimalarial compound artemisinin (Wallaart et al., 2000). Currently, artemisinin is recommended by the World Health Organization for use in combination drug therapies against malaria. Moreover, artemisinin displays few and relatively minor side effects in comparison to the traditionally used quinine. Consequently, there is an increasing demand for artemisinin. Due to the low production of the compound in planta (0,2-0,8% of the dry weight), there are many attempts to produce it in other ways. Chemical synthesis proved to be an impossible strategy due to the complex structure of the molecule and the consequent high costs (Rydén et al., 2007). A recent promising strategy, that may complement plant extraction, is synthetic biology (Bouwmeester et al., 1999; Ro et al., 2006; Rydén et al., 2009; Teoh et al., 2006; Teoh et al., 2007; Wallaart et al., 2001; Zhang et al., 2008). This approach requires identification of the biosynthetic genes in the artemisinin pathway and their expression in a suitable host. Initial biochemical studies were used to identify probable intermediates in the biosynthetic pathway (Bertea et al., 2005; Bouwmeester et al., 1999; Wallaart et al., 1999; Wallaart et al., 1999), which was followed by cloning of the responsible genes (Mercke et al., 2000; Ro et al., 2006; Teoh et al., 2006; Wallaart et al., 2001; Zhang et al., 2008). These efforts led to the identification of the terpene cyclase amorpha-4,11-diene synthase (Amds) which catalyses the first committed step of the biosynthetic pathway of artemisinin (figure 1) (Bouwmeester et al., 1999; Mercke et al., 2000; Wallaart et al., 2001). The second step is catalysed by the cytochrome P450 Cyp71av1 which oxidizes amorpha-4,11-diene at C12 to artemisinic acid in three consecutive steps (figure 1) (Ro et al., 2006, Teoh et al., 2006). Release of the product after the second oxidation step, artemisinic aldehyde, is allowing Dbr2, a carbon double bond reductase, to reduce the 11(13) carbon double bond, thereby producing dihydroartemisinic aldehyde (Zhang et al., 2008).

Chapter 5

128

H

H

Cyp71av1 Cyp71av1H

H

OH

H

H

O

Loss due to chemicalreactivity

Amorpha-4,11-diene Artemisinic alcohol Artemisinic aldehyde

H

H

OH

Cyp71av1 Cyp71av1H

H

O

H

H

O

OH

Dihydroartemisinicalcohol

Dihydroartemisinicaldehyde

Dihydroartemisinicacid

yeast Dbr2

Artemisinin

Less loss due to lowerchemical reactivity

H

H

O

OH

Artemisinic acid

Cyp71av1

Figure 1. Production of dihydroartemisinic acid in yeast during low-oxygen conditions. Cyp71av1 oxidizes amorpha-4,11-diene to artemisinic alcohol, artemisinic aldehyde and artemisinic acid. In A. annua, the carbon double bond reductase Dbr2 reduces artemisinic aldehyde to dihydroartemisinic aldehyde. We show here that in yeast an unidentified enzyme, UDbr, converts artemisinic alcohol to dihydroartemisinic alcohol. We also show that Cyp71av1 can oxidize dihydroartemisinic alcohol to dihydroartemisinic aldehyde and dihydroartemisinic aldehyde to dihydroartemisinic acid, the closest known biosynthetic precursor of artemisinin.

Biochemical studies have indicated that conversion of dihydroartemisinic aldehyde to dihydroartemisinic acid occurs through soluble proteins rather than membrane bound cytochrome P450s (Bertea et al., 2005; Teoh et al., 2006), but the responsible gene has insofar not been identified. In contrast to these reported


129

results, we can show that a membrane bound cytochrome P450 from A. annua is able to efficiently convert dihydroartemisinic aldehyde to dihydroartemisinic acid.

With the availability of all these pathway genes, it should now be possible to create heterologous hosts to produce artemisinin or artemisinin precursors. The focus on artemisinin derived from the plant has in so far not relied so much on genetic engineering, as on growth condition improvement, harvesting methods and marker assisted breeding (Graham et al., 2010; Rydén et al., 2007). Approaches to enhance artemisinin production in A. annua includethe transient expression of a transcription factor, which regulates the expression of amorpha-4,11-diene synthase (Ma et al., 2009). Furthemore, a stable expression of 3-hydroxy-3-methylglutaryl CoA reductase, an enzyme involved in precursor production for general terpene biosynthesis, lead to a 22% increase in artemisinin content compared to control plants (Aquil et al., 2009). Silencing of the squalene biosynthetic pathway, which competes with the artemisinin biosynthetic pathway for precursor molecules, led to a 3 fold increase in artemisinin production (Zhang et al., 2009). As part of a synthetic biology approach, much effort has been spent on improving the precursor-substrate availability in the heterologous microbial hosts. By increasing the production of early precursors that feed into the artemisinin biosynthetic pathway, a greater yield of the desired product may be obtained. For example, the flux through the mevalonate pathway in Escherichia coli (E. coli) (Martin et al., 2003; Pitera et al., 2007; Tsuruta et al., 2009) and Saccharomyces cerevisiae (S. cerevisiae) (Ro et al., 2006) has been optimized. Although hampered by stress related responses (Ro et al., 2008), the use of S. cerevisiae seems promising. By engineering the mevalonate pathway and introducing genes from the artemisinin pathway, a titer of 100 mg l-1 of artemisinic acid was obtained (Ro et al., 2006). Artemisinic acid in its turn can be chemically converted into artemisinin (Ro et al., 2006).

Chapter 5

130

To further increase the production of artemisinin precursors by the heterologous host, it is necessary to reduce any stress factors as much as possible. Expression of heterologous genes introduces stress responses, which will drain resources from artemisinin precursor production. Therefore, it may be desirable to use endogenous genes which are capable of performing the same conversion as genes from the artemisinin pathway in A. annua. Here, we investigated whether S. cerevisiae is able to convert any intermediates in the artemisinin biosynthetic pathway by combining endogenous and heterologous genes. The endogenous metabolic possibilities of S. cerevisiae in artemisinin production were studied by cultivating the host in shake flask cultures with different atmospheric compositions. As gene expression in S. cerevisiae changes in response to alterations of atmospheric composition, it is possible that endogenous genes performing the desired conversions of artemisinin intermediates are up-regulated under such conditions. In addition, we show that Cyp71av1 which is a key enzyme in the artemisinin biosynthetic pathway, has a wider substrate specificity than originally reported (Ro et al., 2006; Teoh et al., 2006).


Microsomal conversion of artemisinin intermediates

As outlined in figure 1, the first committed precursor in the artemisinin biosynthetic pathway is amorpha-4,11-diene, which is oxidized by the cytochrome P450 Cyp71av1 to artemisinic alcohol, artemisinic aldehyde and artemisinic acid (Ro et al., 2006; Teoh et al., 2006). In A. annua, the exocyclic carbon double bond of artemisinic aldehyde is reduced by the reductase Dbr2 (Zhang et al., 2008). The product dihydroartemisinic aldehyde is reportedly further oxidized to dihydroartemisinic acid by cytosolic dehydrogenases (Teoh et al., 2007). Dihydroartemisinic acid is thought to be the closest precursor to artemisinin (Brown et al., 2004). Extracts of A.annua contain other artemisinin-like sesquiterpenoids, that may be


131

intermediates of the artemisinin pathway, but their biosynthesis and involvement in artemisinin formation remain obscure (Lommen et al., 2007; Lommen et al., 2006). Dihydroartemisinic alcohol is one of these compounds that is likely involved in the artemisinin pathway. Previously we have shown that microsomal fractions of A. annuawere able to convert dihydroartemisinic alcohol to dihydroartemisinic aldehyde (Bertea et al., 2005). Due to the close chemical relationship between artemisinic alcohol and dihydroartemisinic alcohol, we hypothesized that Cyp71av1 can also convert dihydroartemisinic alcohol to dihydroartemisinic acid just as it can convert artemisinic alcohol to artemisinic acid. Microsomal fractions of Cyp71av1 expressing S. cerevisiae catalysed the conversion of amorpha-4,11-diene to artemisinic acid as previously published (figure 2a, supplemetary figure 1a) (Ro et al., 2006; Teoh et al., 2006).

a

0

10

20

Tot

al Io

n C

urre

nt

23.00 24.00 25.00 26.00 27.00 28.00 29.00 30.00 31.00Time (min)

CN*

AA

b

0

10

20

30

40

50

60

70

80

90

100

Tot

al Io

n C

urre

nt

23.00 24.00 25.00 26.00 27.00 28.00 29.00 30.00 31.00Time (min)

DHAAA

AMD

0

2

4

6

8

10

Tot

al Io

n C

urre

nt

29.50 30.00 30.50 31.00Time (min)

CN

DHAAOH

AAOH

DHAA

AA

Chapter 5

132

c

0

10

20

30

40

50

60

70

80

90

100

Tot

al Io

n C

urre

nt

23.00 24.00 25.00 26.00 27.00 28.00 29.00 30.00 31.00Time (min)

AMD

CN

AAOH

d

0

10

20

30

40

50

60

Tot

al Io

n C

urre

nt

23.00 24.00 25.00 26.00 27.00 28.00 29.00 30.00 31.00Time (min)

AMD

*CN

e

0

10

20

30

Tot

al Io

n C

urre

nt

24.50 25.00 25.50 26.00 26.50 27.00 27.50 28.00 28.50 29.00 29.50 30.00 30.50 31.00Time (min)

CN

DHAA

f

0

10

20

Tot

al Io

n C

urre

nt

24.50 25.00 25.50 26.00 26.50 27.00 27.50 28.00 28.50 29.00 29.50 30.00 30.50 31.00Time (min)

CN

DHAAA


133

g

0

10

20

Tot

al Io

n C

urre

nt

24.50 25.00 25.50 26.00 26.50 27.00 27.50 28.00 28.50 29.00 29.50 30.00 30.50 31.00Time (min)

CN

DHAAA

DHAAOH

DHAA

h

0

10

20

30

40

50

60

70

Tot

al Io

n C

urre

nt

24.50 25.00 25.50 26.00 26.50 27.00 27.50 28.00 28.50 29.00 29.50 30.00 30.50 31.00Time (min)

CN

DHAAOH

Figure 2. Microsomal conversion of artemisinin intermediates. AMD – amorpha-4,11-diene, AA – artemisinic acid, AAOH – artemisinic alcohol, DHAA – dihydroartemisinic acid, DHAAOH – dihydroartemisinic alcohol, DHAAA – dihydroartemisinic aldehyde, CN – cis-nerolidol, (*)-unknown. a. Undiluted yeast microsomes over-expressing Cyp71av1 with AMD as substrate. b. 10x diluted yeast microsomes over-expressing Cyp71av1 with AMD as substrate. c. 100x diluted yeast microsomes over-expressing Cyp71av1 with AMD as substrate. d. Empty vector control: Undiluted yeast microsomes with AMD as substrate. e. Undiluted yeast microsomes over-expressing Cyp71av1 with DHAAA as substrate. f. Empty vector control: Undiluted yeast microsomes with DHAAA as substrate. g. Undiluted yeast microsomes over-expressing Cyp71av1 with DHAAOH as substrate. h. Empty vector control: Undiluted yeast microsomes with DHAAOH as substrate.

Chapter 5

134

Furthermore, we were able to show that microsomes containing Cyp71av1 oxidized dihydroartemisinic alcohol and dihydroartemisinic aldehyde to dihydroartemisinic acid (figure 2e-h, supplementary figure 1e-h). It is thus theoretically possible that there is an additional pathway proceeding from artemisinic alcohol to dihydroartemisinic alcohol, rather than from artemisinic aldehyde to dihydroartemisinic aldehyde (figure 1). The conversion rates of artemisinic alcohol to artemisinic acid and dihydroartemisinic alcohol to dihydroartemisinic acid do not seem to differ significantly, as the final concentrations of the acids in the samples are similar in the 10x diluted microsomal fractions supplemented with amorpha-4,11-diene as substrate (table 1). However an exact kinetic analysis is difficult to conduct due to the mixed protein nature of microsomes and the losses of artemisinic aldehyde (table 1). Generally aldehydes are more reactive than alcohols, and it is likely that artemisinic aldehyde and dihydroartemisinic aldehyde, either supplied as substrate or formed enzymatically from earlier precursors, bind more to the microsomal fraction than artemisinic alcohol and dihydroartemisinic alcohol. The reactivity of artemisinic aldehyde is illustrated by the absence of the compound in the 10-fold diluted microsomal fractions with amorpha-4,11-diene as substrate, while all other intermediates up to artemisinic acid were detected. Furthermore, in the undiluted microsomal fractions which were supplemented with dihydroartemisinic aldehyde as substrate, it could not be detected anymore after the assay. The loss of artemisinic aldehyde is expected to be even more severe as it is chemically more reactive than dihydroartemisinic aldehyde.

Another interesting unexpected result is the detection of dihydroartemisinic alcohol, dihydroartemisinic aldehyde and dihydroartemisinic acid in a microsomal conversion assay with amorpha-4,11-diene as substrate (figure 2b, supplementary figure 1b). Previously, only the conversion products artemisinic alcohol, artemisinic aldehyde and artemisinic acid have been detected in


135

Table 1. Retention times and abundances of the internal standard cis-nerolidol and produced artemisinin intermediates in the microsomal assays. Dilution series of microsomes were mixed with amorpha-4,11-diene as substrate. Undiluted microsomes were incubated with either amorpha-4,11-diene, dihydroartemisinic aldehyde or dihydroartemisinic alcohol.

Chemical component Retention time [min]

Microsomal fractions

Undiluted 10x dilution 100x dilution

Terpenoid [µM]

Terpenoid [µM]

Terpenoid [µM]

Internal standard

Cis-nerolidol 24.2 15 15 15

Metabolic intermediates, AMD as substrate

Amorpha-4,11-diene 23.3 - 2.6 5.7

Artemisinic alcohol 29.2 - 2.8 4.0

Artemisinic acid 31.0 2.2 2.6 -

Dihydroartemisinic alcohol 29.1 - 2.7 -

Dihydroartemisinic aldehyde 27.3 - 3.0 -

Dihydroartemisinic acid 30.3 - 2.6 -

Metabolic intermediates, DHAAA as substrate

Dihydroartemisinic acid 30.3 2.6 - -

Metabolic intermediates, DHAAOH as substrate

Dihydroartemisinic alcohol 29.1 2.7 - -

Dihydroartemisinic aldehyde 27.3 2.1 - -

Dihydroartemisinic acid 30.3 2.3 - -

Chapter 5

136

microsomal assays using amorpha-4,11-diene as substrate (Teoh et al., 2006). This suggests that S. cerevisiae microsomal preparations contain an unspecific 11(13) carbon double bond reductase (Udbr). The fact that both artemisinic acid and dihydroartemisinic acid are produced (figure 2b) suggests that competition occurs between Cyp71av1 and Udbr for the substrate artemisinic alcohol. In undiluted microsomal fractions, a complete conversion of amorpha-4,11-diene to artemisinic acid is observed (table 1). When microsomes are diluted 100-fold, the enzymes involved become a limiting factor and only artemisinic alcohol is produced from amorpha-4,11-diene. However, in a 10-fold diluted microsomal preparation, there is a window for production of the full range of artemisinin intermediates. This is probably explained by different kinetic properties of Cyp71av1 and Udbr. The results infer that a sufficient amount of artemisinic alcohol needs to accumulate before Cyp71av1 can recapture the molecules and further oxidize them to artemisinic aldehyde. Again, a minimal amount of artemisinic aldehyde must be reached before artemisinic aldehyde is oxidized to artemisinic acid. These enzyme-intermediate capture-release cycles create an opportunity for other non-specific enzymes, such as Udbr, to act on the molecules. Hence, as artemisinic alcohol accumulates, part of it will be reduced to dihydroartemisinic alcohol by Udbr and part of it will be oxidized to artemisinic aldehyde by Cyp71av1. However, the affinity of Cyp71av1 for artemisinic alcohol is greater than that of Udbr, as is evident from the 100x microsomal dilution assay, where only artemisinic alcohol is detected but no dihydroartemisinic alcohol. If the reverse situation would be true, that the affinity of Udbr for artemisinic alcohol would be greater than that of Cyp71av1, then we would have detected dihydroartemisinic alcohol in the sample.

Shake flask cultures

To confirm the relevance of Udbr in production of dihydroartemisinic acid in vivo, S. cerevisiae WAT11 was


137

transformed with cyp71av1 and intact cells were incubated with amorpha-4,11-diene as substrate in shake flask cultures (figure 3, table 2, supplementary figure 2).

a

0

10

20

30

40

50

60

70

80

90

100

Tot

al Io

n C

urre

nt

23.00 24.00 25.00 26.00 27.00 28.00 29.00 30.00 31.00Time (min)

CN

AMD DHAA AA

DHAAOH

b

0

10

20

30

40

50

60

70

80

90

100

Tot

al Io

n C

urre

nt

23.00 24.00 25.00 26.00 27.00 28.00 29.00 30.00 31.00Time (min)

AMD

CN

DHAAOHDHAA

AA

c

0

10

20

30

40

50

60

70

80

90

100

Tot

al Io

n C

urre

nt

23.00 24.00 25.00 26.00 27.00 28.00 29.00 30.00 31.00Time (min)

CN DHAAOH

DHAA

AA

d

0

10

20

30

40

50

60

70

Tot

al Io

n C

urre

nt

23.00 24.00 25.00 26.00 27.00 28.00 29.00 30.00 31.00Time (min)

CNDHAAOH

DHAAAA

Chapter 5

138

Figure 3. Shake flask cultures with S. cerevisiae WAT11. AMD – amorpha-4,11-diene,, AA – artemisinic acid, DHAA – dihydroartemisinic acid, DHAAOH – dihydroartemisinic alcohol, CN – cis-nerolidol.

a. S. cerevisiae WAT11 over-expressing Cyp71av1. Incubation in shaking flask with nitrogen atmosphere. AMD was added as substrate. b. S. cerevisiae WAT11 over-expressing Cyp71av1. Incubation in shaking flask with atmosphere identical to normal air composition. AMD was added as substrate. c. S. cerevisiae WAT11 over-expressing Cyp71av1. Incubation in shaking flask with loose cotton plug. AMD was added as substrate. d. S. cerevisiae WAT11 over-expressing Cyp71av1. Incubation in shaking flask with tight cotton plug. AMD was added as substrate.

Table 2. Production of artemisinin intermediates in shake flask cultures with varying atmospheric compositions.

Culture conditions AMD[mg/L]

DHAAOH [mg/L]

DHAA[mg/L]

AA [mg/L]

Shaking flask, nitrogen 5.5-6.1 10.1-10.9 6.7-7.4 6.6-7.5

Shaking flask, aired 4.2-4.9 7.2-7.8 6.0-6.5 6.9-7.6

Shaking flask, tight cotton plug - 4.6-5.3 4.7-5.0 4.4-4.8

Shaking flask, loose cotton plug - 3.9-4.7 4.6-5.1 4.8-5.4

To further test the hypothesis of enzymatic competition we included oxygen as a variable parameter, since this is expected to affect the efficiency of Cyp71av1 but not of Udbr. Moreover, gene expression in S. cerevisiae is known to drastically change with variations in atmospheric oxygen level (Rintala et al., 2009). The number of genes affected by oxygen limitation is also depending on the carbon source where galactose has been shown to have a stronger impact on the transcriptome than glucose (Lai et al., 2006). The results show a


139

trend towards higher titers of artemisinic acid than dihydroartemisinic acid in cultures that have more oxygen (table 2). This suggests that Cyp71av1 catalysis is less efficient under low oxygen availability and/or that the expression of Udbr is increased under these conditions. The volatile nature of the sesquiterpenes is evident as the amounts of artemisinin intermediates are significantly lower in flasks sealed solely with cotton. However, the trend of a higher titer of dihydroartemisinic acid compared to artemisinic acid in limited oxygen conditions is preserved.


To confirm the influence of oxygen on dihydroartemisinic acid production, S. cerevisiae WAT 11 expressing Amds and Cyp71av1 was grown under different atmospheric conditions by flushing the cultures with normal air or air mixed with nitrogen. Maximum oxygen level (100%) was defined as normal air entering the culture at a rate of 250 ml min under constant stirring at 350 rpm. 20% of this oxygen value was defined as low level oxygen condition and was achieved by mixing in nitrogen with air. Thus, an experimental setup with high versus low oxygen levels was created. In these fermentor studies, S. cerevisiae WAT11 transformed with amds and cyp71av1 yielded 24 mg L-1 dihydroartemisinic alcohol in low level

0

10

20

30

40

50

60

70

80

90

100

Tot

al Io

n C

urre

nt

23.00 24.00 25.00 26.00 27.00 28.00 29.00 30.00 31.00Time (min)

AMD

CN

DHAAOH

DHAA

Figure 4. Fermentor study with S. cerevisiae WAT11 over-expressing Amds and Cyp71av. Sampling after 2 days incubation. AMD – amorpha-4,11-diene, AA – artemisinic acid, DHAA – dihydroartemisinic acid, DHAAOH – dihydroartemisinic alcohol.

Chapter 5

140

Table 4. Production of DHAAOH in fermentor cultures.

Sample DHAAOH[mg/L]

Cyp71av1, low oxygen concentration 24-25.9

Cyp71av1, high oxygen concentration 8.1-8.9

oxygen cultures (figure 4, table 3, supplementary figure 3). In contrast, fermentations performed in oxygen rich environment produced three times less dihydroartemisinic aldehyde (8 mg L-1). As discussed above, the artemisinin intermediates are volatile and the lack of artemisinic acid and dihydroartemisinic acid in the fermentor samples may be explained by a combination of low production and escape of volatile intermediates with waste air. Therefore dihydroartemisinic alcohol, which is the first precursor to dihydroartemisinic acid, was accepted as an indicator for the capacity of dihydroartemisinic acid production. An attempt to trap the volatiles in an organic overlay of dodecane resulted in no detectable production of either artemisinic acid or dihydroartemisinic acid, but rather the accumulation of amorpha-4,11-dinene in the organic phase (data not shown).

Concluding remarks

In this communication we demonstrate for the first time the conversion of dihydroartemisinic alcohol to dihydroartemisinic aldehyde and dihydroartemisinic acid by Cyp71av1 (figure 2e-h; figure 3). We exploit this feature in a fermentation study and report that under conditions with limited oxygen concentration and galactose as sole carbon source, S. cerevisiae transformed with amds and cyp71av1 is capable of reducing the early artemisinin intermediate artemisinic alcohol to dihydroartemisinic alcohol (figure 1). The corresponding enzyme has up to date not been found in A. annua.


141

Imposed heterologous genes impart stress on the host, and the stress reactions drain resources from production of the desired metabolite. One way of limiting the stress is to use as few heterologous genes as possible and combine the biosynthetic pathway with adequate endogenous enzymes instead. It is desirable to choose a host that contains genes compatible with the heterologous biosynthetic pathway. The finding that S. cerevisiae already expresses an aspecific carbon double bond reductase that performs a key step in the artemisinin biosynthesis, may open up possibilities to use one heterologous gene less in microbial artemisinin precursor production. The identity of Udbr remains obscure. Even so, it can be argued that Udbr belongs to a different protein family than Dbr2 (Zhang et al., 2008). Although the substrates of Dbr2 and Udbr are similar, the enzymes act on different chemical groups. Dbr2 reduces the exocyclic carbon double bond of artemisinic aldehyde, thereby producing dihydroartemisinic aldehyde, while Udbr reduce the ketone of dihydroartemisinic aldehyde to an alcohol. Dbr2 is classified into the same protein family as the yeast old yellow enzyme due to its sequence homology and its conformity with the general substrate pattern of , -unsaturated carbonyls (Williams et al., 2002). As Udbr is reducing the carbonyl double bond itself and not a distand carbon-carbon double bond, it is unlikely that it belongs to this vast family of proteins. Regardless of the identity of Udbr, the results show that it may be rewarding to thoroughly investigate the selection of hosts and growth conditions in a synthetic biology project. Some hosts may offer more benefits than others in terms of endogenous promiscuous enzymatic activities.


Chemicals

All chemicals were purchased from Sigma or Difco. DNA polymerase was purchased from Finnzymes and restriction enzymes

Chapter 5

142

from New England Biolabs. Artemisinin intermediates were synthesized as described by Bertea el al. (2005).

Media

SGal: 1% w/v casamino acids, 2% w/v galactose, 0.67% w/v yeast nitrogen base with ammonium sulfate without amino acids. SGlu: 1% w/v casamino acids, 2% w/v glucose, 0.67% w/v yeast nitrogen base with ammonium sulfate without amino acids. Agar plates were prepared by adding 1.5% w/v agar to the medium. YPGA: 1% w/v yeast extract, 1% w/v bactopeptone, 2 % w/v glucose, 200 mg l-1

adenine. YPL: 1% w/v yeast extract, 1% w/v bactopeptone, 2% w/v galactose.

Strains and plasmids

Escherichia coli (E.coli) XL-1 Blue (recA1 endA1 gyrA96 thi-1 hsdR17 supE44 relA1 lac [F´ proAB lacIqZ M15 Tn10 (TetR)])(Stratagene) and the vector pGEMT (Promega) was used for molecular cloning and sequencing. The construction of the Saccharomyces cerevisiae (S. cerevisiae) strain WAT11, a derivative of the W303-B strain (MAT a; ade2–1; his3–11,-15; leu2–3,-112;ura3–1; canR; cyr+) expressing the ATR1 Arabidopsis thalianaNADPH–P450 reductase has been described (Truan et al., 1993; Urban et al., 1997). S. cerevisiae WAT11 was used as destination host and the vectors pYeDP60 and pYeDP80 for protein expression (Pompon et al., 1996). The episomal high copy expression vectors pYeDP60 and pYeDP80 are induced by galactose and confer adenine and tryptophane or uracil auxotrophy.

Cloning of amds and cyp71av1

Amds was a gift from Mattijs Julsing (Department of Pharmaceutical biology, University of Groningen). Primers 1 and 2 were used to


143

amplify amds with the added restriction sites BamHI and EcoRI (table 5).

Table 5. List of primers used in the study. Underlined nucleotides reprecent restriction sites.

No. primer Sequence 5’ to 3’ Restriction site

1 5’ gggatccatgtcacttacagaagaaaaacc 3’ Bam HI

2 5’ ggaattctcatatactcataggataaacgag 3’ EcoRI

3 5’ cggcggatccatggcactctcactgaccacttcca 3’ Bam HI

4 5’ cggcggtaccctagaaacttggaacgagtaacaactcagcctttc 3’ KpnI

The reaction conditions recommended by the manufacturer of Phusion DNA polymerase (Finnzymes) were followed using the PCR program: Initial denaturation for 1 min at 98°C, 30 cycles at 98°C for 10 sec, 70°C for 20 sec, 72°C for 30 sec and a final extension step at 72°C for 10 min. The construct was cloned into pGEMT (Promega), propagated in E. coli XL1-Blue and sequenced. A confirmed construct was subcloned from pGEMT by restriction with BamHI and EcoRI and ligated into the expression vector pYeDP80. The construct pYeDP80-amds was transformed into E.coli XL1-Blue and confirmed by restriction analysis.

EST sequences of a glandular trichome specific cDNA library (Bertea et al., 2006) were compared with Genbank sequences using batch BlastX (NCBI) with standard settings. Interesting candidate cytochrome P450s were extended to full length using the EST library described above. The full length gene of cyp71av1 (Ro et al., 2006; Teoh et al., 2006), was obtained and subcloned into the sequencing vector pGEMT using the primers 3 and 4 listed in table 1. Cyp71av1 was picked up from the cDNA library by following the reaction conditions recommended by the manufacturer of Phusion DNA polymerase (Finnzymes) and the PCR program: Initial denaturation for 1 min at 98°C, 30 cycles at 98°C for 10 sec, 70°C for 20 sec,

Chapter 5

144

72°C for 30 sec and a final extension step at 72°C for 10 min. The construct was propagated in E. coli XL1-Blue and confirmed with sequencing. Thereafter, cyp71av1 was restricted from pGEMT using BamHI and KpnI and ligated into the expression vector pYeDP60. The construct pYeDP60-cyp71av1 was transformed into E. coli XL1-Blue and confirmed by restriction analysis.

Microsomal isolation and conversion assays

The constructs pYeDP60 and pYeDP60-cyp71av1 were transformed into S. cerevisiae WAT11 using the LiAc method (Maniatis et al., 1982). Mutants were selected on SGal agar plates supplemented with 100 μg ml-1 ampicillin. Colonies were checked for the presence of the constructs by yeast colony PCR (Sambrook et al., 2001). Small scale cultures of 50 ml SGal were initiated with S. cerevisiae WAT11 containing either pYeDP60 or pYeDP60-cyp71av1. S. cerevisiae WAT11 transformed with the empty vector was used as negative control. The seed cultures were incubated for 40 hours at 28°C with shaking. Following incubation the cultures were centrifuged and the pellets were resuspended in 250 ml YPL medium. Microsomes were isolated using the method described by Pompon et al. (1996) and stored in 200 μl aliquots at -80°C. Protein content was estimated to 10.7-10.9 mg ml-1 in Bradford assays using bovine serum albumin as standard (Biorad). Conversion assays were performed using 200 μl microsomes in 800 μl assay buffer (50 mM Tris-HCl pH 7.4, 1 mM EDTA, 1mM DTT, 4% w/v glycerol, 5 μM FAD, 5 μM FMN, 2 mM glucose 6-phosphate, 1 unit glucose 6-phosphate dehydrogenase, 1mM -NADPH and 30 μM amorpha-4,11-diene, dihydroartemisinic aldehyde or dihydroartemisinic alcohol. Dilution series contained 2.1 mg, 0.21 mg and 21 μg microsomal protein. Assays were incubated for 2 hours at 30°C with gentle shaking. Reactions were stopped on ice. 15 μM Cis-nerolidol was added as internal standard. The assay was then extracted 3 times with 1 ml ethyl acetate followed by a final extraction with 1.5 ml ethyl acetate. The combined organic extractions were dried with


145

anhydrous Na2SO4 and evaporated to 200 μl under a gentle flow of N2. Samples were analyzed by GC-MS as described below. Peaks were identified using authentic standards. Product yields were calculated based on the internal standard peak area and calibration curves made from authentic standards.

Shake flask studies

5 ml aliquots of SGlu were inoculated with appropriate combinations of S. cerevisiae WAT11 pYeDP60, pYeDP80, pYeDP60-cyp71av1or pYeDP80-amds. The seed cultures were incubated at 30°C with gentle shaking until an OD600 of 0.5 was reached. Main cultures were initiated by adding seed culture to a calculated final OD600 of 0.05 to 50 ml SGal in 250 ml shake flasks with airtight glass stoppers or cotton plugs. The atmosphere was adjusted by adding 350 ml l-1 air or nitrogen gas for 1 min to flasks sealed with glass stoppers. The atmosphere of these cultures was refreshed by flushing the flasks for 2 min with 350 ml l-1 air or nitrogen after 24 h of incubation. Cultures sealed with loose or tight cotton plugs were used without any atmospheric modification. All cultures were incubated at 30°C with shaking for 48 h. Thereafter the cultures were cooled for 30 min at 4°C and extracted with 15 ml ethyl acetate. To 2 ml of this organic phase, 0.83 μg cis-nerolidol was added as internal standard. The samples were filtered and dried using a small column containing silica and anhydrous Na2SO4 which was eluted with an additional 1 ml of ethyl acetate. Before GC-MS analysis, as described below, the samples were concentrated to approximately 500 μl under a gentle flow of N2. Product yields were calculated based on the internal standard peak area and calibration curves made from authentic standards.


100 ml SGlu seed cultures were initiated by inoculation with appropriate combinations of S. cerevisiae WAT11 pYeDP60,

Chapter 5

146

pYeDP80, pYeDP60-cyp71av1 or pYeDP80-amds. Cultures were incubated at 30°C with shaking until OD600 4.0 was reached. Bench top fermentors (New Brunswick Scientific One BioFlo 310) with a total volume of 5 l were used in the fermentation studies. The level of dissolved oxygen (DO2) was calibrated to 100% in the medium at 30°C by stirring at 350 rpm and adding air at a flow of 250 ml min-1.Fermentor cultures were started by adding seed cultures to a calculated value of OD600 of 0.05 in 3 l SGal. By mixing in nitrogen with air as well as automatically adjusting the stirring speed, DO2was kept at either 100% or 20% of the calibrated initial value. Thus, an environment was created which was either rich in oxygen (100% DO2) or low in oxygen (20% DO2) content. Incubation was maintained for 48 hours. From the cultures 55 ml samples were extracted with 30 ml ethyl acetate spiked with 15 μM of the internal standard cis-nerolidol. The samples were subsequently extracted twice more with 30 ml ethyl acetate. The organic phases were combined, and the water removed with anhydrous Na2SO4. The extracts were concentrated in a vacuum centrifuge to approximately 8 ml. A small aliquot was analyzed by GC-MS, as described below. Product yields were calculated based on the internal standard peak area and calibration curves made from authentic standards.

GC-MS

GC-MS analysis was performed on an Agilent 6890 series gas chromatograph coupled to a JEOL JMS-SUN 200 mass selective detector in electron impact mode (70 eV) equipped with a guard column (1 m x 0.25 mm i.d., GL-Sciences) and a HP-5MS analytical column (30 m x 0.25 mm i.d., 0.25 μm film). The temperature program was set to an initial temperature of 60°C followed by a first temperature gradient of 3°C min-1 to 220°C and then a second gradient of 30°C min-1 to 300°C with an endpoint hold of 4 min. Helium flow was set to 0.7 ml min-1, detector temperature and injector temperature to 250°C using 500 scans min-1 with a scan range of 40-350 m/z. A persistent problem with peak tailing of


147

dihydroartemisinic alcohol in in vivo assays could not be solved, but this did not hamper reliable peak integration.

References

Aquil, S., Husaini, A. M., Abdin, M. Z., & Rather, G. M. (2009). Overexpression of the HMG-CoA reductase gene leads to enhanced artemisinin biosynthesis in transgenic Artemisia annua plants. Planta Med, 75(13), 1453-1458.





Graham, I. A., Besser, K., Blumer, S., Branigan, C. A., Czechowski, T., Elias, L., et al. (2010). The Genetic Map of Artemisia annua L. Identifies Loci Affecting Yield of the Antimalarial Drug Artemisinin. Science, 327(5963), 328-331.

Lai, L. C., Kosorukoff, A. L., Burke, P. V., & Kwast, K. E. (2006). Metabolic-state-dependent remodeling of the transcriptome in response to anoxia and subsequent reoxygenation in Saccharomyces cerevisiae. Eukaryot Cell, 5(9), 1468-1489.

Chapter 5

148

Lommen, W. J., Elzinga, S., Verstappen, F. W., & Bouwmeester, H. J. (2007). Artemisinin and sesquiterpene precursors in dead and green leaves of Artemisia annua L. crops. Planta Med, 73(10), 1133-1139.


Ma, D., Pu, G., Lei, C., Ma, L., Wang, H., Guo, Y., et al. (2009). Isolation and Characterization of AaWRKY1, an Artemisia annua Transcription Factor that Regulates the Amorpha-4,11-diene Synthase Gene, a Key Gene of Artemisinin Biosynthesis. Plant Cell Physiol., 50(12), 2146-2161.

Maniatis, T., Fritsch, E. F., & Sambrook, J. (1982). Molecularcloning: a laboratory manual: Cold Spring Harbor, N.Y. : Cold Spring Harbor Laboratory.

Martin, V. J., Pitera, D. J., Withers, S. T., Newman, J. D., & Keasling, J. D. (2003). Engineering a mevalonate pathway in Escherichia coli for production of terpenoids. NatBiotechnol, 21(7), 796-802.


Pitera, D. J., Paddon, C. J., Newman, J. D., & Keasling, J. D. (2007). Balancing a heterologous mevalonate pathway for improved isoprenoid production in Escherichia coli. Metab Eng, 9(2),193-207.

Pompon, D., Louerat, B., Bronine, A., & Urban, P. (1996). Yeast expression of animal and plant P450s in optimized redox environments. Methods Enzymol, 272, 51-64.

Rintala, E., Toivari, M., Pitkanen, J. P., Wiebe, M. G., Ruohonen, L., & Penttila, M. (2009). Low oxygen levels as a trigger for


149

enhancement of respiratory metabolism in Saccharomyces cerevisiae. BMC Genomics, 10, 461.

Ro, D. K., Ouellet, M., Paradise, E. M., Burd, H., Eng, D., Paddon, C. J., et al. (2008). Induction of multiple pleiotropic drug resistance genes in yeast engineered to produce an increased level of anti-malarial drug precursor, artemisinic acid. BMC Biotechnol, 8, 83.


Rydén, A.-M., & Kayser, O. (2007). Sesquiterpene trioxane lactones - chemistry and biological activities Topics in heterocyclic chemistry (Vol. 9, pp. 1-28): Springer.

Rydén, A.-M., Ruyter-Spira, C., Quax, W. J., Osada, H., Muranaka, T., Kayser, O., et al. (2009). Molecular cloning of dihydroartemisinic aldehyde reductase and its implication in artemisinin biosynthesis. In press, Applied microbiology and biotechnology.

Sambrook, J., & Russell, D. W. (2001). Molecular cloning : a laboratory manual (3rd ed.). Cold Spring Harbor, N.Y.: Cold Spring Harbor Laboratory Press.


Teoh, K. H., Reed, D. W., Polichuk, D. R., & Covello, P. S. (2007). W. i. p. organization.

Truan, G., Cullin, C., Reisdorf, P., Urban, P., & Pompon, D. (1993). Enhanced in vivo monooxygenase activities of mammalian P450s in engineered yeast cells producing high levels of NADPH-P450 reductase and human cytochrome b5. Gene,125(1), 49-55.

Chapter 5

150

Tsuruta, H., Paddon, C. J., Eng, D., Lenihan, J. R., Horning, T., Anthony, L. C., et al. (2009). High-level production of amorpha-4,11-diene, a precursor of the antimalarial agent artemisinin, in Escherichia coli. PLoS One, 4(2), e4489.

Urban, P., Mignotte, C., Kazmaier, M., Delorme, F., & Pompon, D. (1997). Cloning, yeast expression, and characterization of the coupling of two distantly related Arabidopsis thaliana NADPH-cytochrome P450 reductases with P450 CYP73A5. J Biol Chem, 272(31), 19176-19186.





Williams, R. E., & Bruce, N. C. (2002). 'New uses for an Old Enzyme' - the Old Yellow Enzyme family of flavoenzymes. Microbiology, 148(6), 1607-1614.

Zhang, L., Jing, F., Li, F., Li, M., Wang, Y., Wang, G., et al. (2009). Development of transgenic Artemisia annua (Chinese wormwood) plants with an enhanced content of artemisinin,


151

an effective anti-malarial drug, by hairpin-RNA-mediated gene silencing. Biotechnol Appl Biochem, 52(Pt 3), 199-207.


Chapter 5

152

Supplementary figures

a

m/z010203040

5060708090

100

AB

UN

DA

NC

E

Scan No : 3941 Ret Time : 31:6.69 TIC : 59821 BP = 5573

235219206189

178173161145136

131

121

115

105

93

87

79

67

61

55

5145

60 80 100 120 140 160 180 200 220

Artemisinic acid

b

m/z010203040

5060708090

100

AB

UN

DA

NC

E


221

202

189

173

159

145132

121

115

105

9179

67

55

51

45

> 50 100 150 200 250

Artemisinic alcohol

m/z010203040

5060708090

100

AB

UN

DA

NC

E


232

204

197

189

175

162147

133

128

119

105

9179

67

61

55

5146

60 80 100 120 140 160 180 200 220

Amorpha-4,11-diene

m/z010203040

5060708090

100

AB

UN

DA

NC

E


267

234

224207

189

178162

147136

121

115

105

9379

67

6155

51

45

> 50 100 150 200 250

Artemisinic acid


153

m/z010203040

5060708090

100

AB

UN

DA

NC

E


222191

175

163

149135

121

115

1079379

67

61

55

51

> 40 60 80 100 120 140 160 180 200 220


m/z010203040

5060708090

100

AB

UN

DA

NC

E


163

147

133121

1059179

6755

51

50 100 150 200 250

Dihydroartemisinic aldeyde

m/z010203040

5060708090

100

AB

UN

DA

NC

E


236207

162

147

133121107

91

7967

55

51

45

> 50 100 150 200 250


c

m/z010203040

5060708090

100

AB

UN

DA

NC

E


205

189

175

162147133

119

105

93

79

67

63

55

51

50 100 150 200 250

Amorpha-4,11-diene

m/z010203040

5060708090

100

AB

UN

DA

NC

E


220

202

189

173

159

145132

128

121

115

105

9179

67

63

55

51

60 80 100 120 140 160 180 200 220

Artemisinic alcohol

Chapter 5

154

d

m/z010203040

5060708090

100

AB

UN

DA

NC

E


205

189

175

162147

133

119

105

93

79

67

63

55

5145

50 100 150 200 250

Amorpha-4,11-diene

e

m/z010203040

5060708090

100

AB

UN

DA

NC

E


162

147

133121

1079179

746755

5145

50 100 150 200 250


f

m/z010203040

5060708090

100

AB

UN

DA

NC

E


162

147

133121115

10591796761

55

5145

50 100 150 200 250


g

m/z010203040

5060708090

100

AB

UN

DA

NC

E


162

147

1331211059179

676155

5145

50 100 150 200 250



155

m/z010203040

5060708090

100

AB

UN

DA

NC

E


222191

163

149135121

1079381

7767

55

51

50 100 150 200 250


m/z010203040

5060708090

100

AB

UN

DA

NC

E


162

147

133121

1079179

746755

45

60 80 100 120 140 160 180 200 220 240


h

m/z010203040

5060708090

100

AB

UN

DA

NC

E


222191

163

149135121

1079381

7767

55

51

50 100 150 200 250


Chapter 5

156

Supplementary figure 1.Mass spectra for GC chromatograms from yeast microsomal assays.

a. Undiluted yeast microsomes over-expressing Cyp71av1 with amorpha-4,11-diene as substrate. b. 10x diluted yeast microsomes over-expressing Cyp71av1 with amorpha-4,11-diene as substrate. c. 100x diluted yeast microsomes over-expressing Cyp71av1 with amorpha-4,11-diene as substrate. d. Empty vector control: Undiluted yeast microsomes with amorpha-4,11-diene as substrate. e. Undiluted yeast microsomes over-expressing Cyp71av1 with dihydroartemisinic aldehyde as substrate. f. Empty vector control: Undiluted yeast microsomes with dihydroartemisinic aldehyde as substrate. g. Undiluted yeast microsomes over-expressing Cyp71av1 with dihydroartemisinic alcohol as substrate. h. Empty vector control: Undiluted yeast microsomes with dihydroartemisinic alcohol as substrate.


157

a

m/z010203040

5060708090

100

AB

UN

DA

NC

E


204

189

175

162

149133

119

115

107

103

9179

67

55

51

41

38

40 60 80 100 120 140 160 180 200

Amorpha-4,11-diene

m/z010203040

5060708090

100

AB

UN

DA

NC

E


236

162

147

133121

1079179

74675541

40 60 80 100 120 140 160 180 200 220


m/z010203040

5060708090

100

AB

UN

DA

NC

E


235216206189173

162147136

131

121

115

105

93

87

79

67

63

55

51

41

40 60 80 100 120 140 160 180 200 220

Artemisinic acid

b

m/z010203040

5060708090

100

AB

UN

DA

NC

E


204189

162149138133

119

115

107

103

9377

67

6360

5551

41

40 60 80 100 120 140 160 180 200

Amorpha-4,11-diene

Chapter 5

158

m/z010203040

5060708090

100

AB

UN

DA

NC

E


222191

163

147135121

115

1079379

67

55

51

41

40 60 80 100 120 140 160 180 200 220


m/z010203040

5060708090

100

AB

UN

DA

NC

E


236

162

147

133121

1079179

746755

51

41

40 60 80 100 120 140 160 180 200 220


m/z010203040

5060708090

100

AB

UN

DA

NC

E


235219206201

189178173

162147136

131

121

115

105

93

87

79

67

63

55

51

41

40 60 80 100 120 140 160 180 200 220

Artemisinic acid

c

m/z010203040

5060708090

100

AB

UN

DA

NC

E


222191

163

147135121

107

9381

7769

55

51

41

40 60 80 100 120 140 160 180 200 220


m/z010203040

5060708090

100

AB

UN

DA

NC

E


191

162

147

135121

115

1079179

67

55

51

41

40 60 80 100 120 140 160 180 200 220

Dihydrartemisinic acid


159

m/z010203040

5060708090

100

AB

UN

DA

NC

E


235216206201

189173161145

136

131

121

115

105

93

87

79

67

63

55

51

41

40 60 80 100 120 140 160 180 200 220

Artemisinic acid

d

m/z010203040

5060708090

100

AB

UN

DA

NC

E


222191

163

149135121

107

9381

7769

55

41

40 60 80 100 120 140 160 180 200 220


m/z010203040

5060708090

100

AB

UN

DA

NC

E


236

162

147

133121

107917967

5541

40 60 80 100 120 140 160 180 200 220


m/z010203040

5060708090

100

AB

UN

DA

NC

E


235216206189

173

161147

136130

121

115

107

93

87

79

67

63

55

51

41

40 60 80 100 120 140 160 180 200 220

Artemisinic acid

Chapter 5

160

Supplementary figure 2.Mass spectra for GC chromatograms from shake flask cultures.

a. S. cerevisiae WAT11 over-expressing Cyp71av1. Incubation in shaking flask with nitrogen atmosphere. Amorpha-4,11-diene was added as substrate. b. S. cerevisiae WAT11 over-expressing Cyp71av1. Incubation in shaking flask with atmosphere identical to normal air composition. Amorpha-4,11-diene was added as substrate. c. S. cerevisiae WAT11 over-expressing Cyp71av1. Incubation in shaking flask with loose cotton plug. Amorpha-4,11-diene was added as substrate. d. S. cerevisiae WAT11 over-expressing Cyp71av1. Incubation in shaking flask with tight cotton plug. Amorpha-4,11-diene was added as substrate.


161

m/z010203040

5060708090

100

AB

UN

DA

NC

E


205

189

175

162147138

133

128

119

115

107

103

93

89

79

67

63

55

51

41

40 60 80 100 120 140 160 180 200

Amorpha-4,11-diene

m/z010203040

5060708090

100

AB

UN

DA

NC

E


222

202

189

173

163

145132

121

115

107

9379

67

63

55

51

41

40 60 80 100 120 140 160 180 200 220


m/z010203040

5060708090

100

AB

UN

DA

NC

E


222

162

147

135130

121

1079181

7767

5541

40 60 80 100 120 140 160 180 200 220


Supplementary figure 3.Mass spectra for GC chromatograms from fermentor study.

Chapter 5

162

Chapter 6

Role of peroxygenase in artemisinin production

Anna-Margareta Rydén1, 2, Toshiya Muranaka2,Wim J. Quax1, Oliver Kayser1

1 Department of Pharmaceutical biology, University of Groningen, Groningen 9713 AV, the Netherlands 2Kihara Institute for Biological Research, Yokohama City University, 641-12, Maioka-cho, Tostuka-ku, Yokohama 244-0813, Japan

Chapter 6

164

Abstract

Artemisinin is an antimalarial drug which is produced by the medicinal plant Artemisia annua L. (A. annua). The yields are notoriously low (0,2-0,8% dry weight), and therefore several approaches to increase the production are being explored. One strategy is focused on elucidating the biosynthetic pathway of artemisinin. By understanding the biosynthesis of the drug, and by isolating the responsible genes, it will be possible to produce it in heterologous hosts. Currently, the genes involved in the production of the early intermediates are known: production of artemisinic acid and dihydroartemisinic acid in yeast has been reported. These intermediates can be chemically converted to artemisinin. However, it is desirable to directly obtain artemisinin as end product in heterologous production platforms. In order to satisfy this condition, an understanding of what happens with the the intermediates artemisinic acid and dihydroartemisinic acid is needed. It is not known how these intermediates are turned into artemisinin in planta.We here investigate the enzymatic involvement in formation of late intermediates in the artemisinin biosynthetic pathway.

Peroxygenase in A. annua

165

Introduction

Artemisinin is produced in small quantities (0.01-2% dry weight (Zhang et al., 2008)) in the plant Artemisia annua L. (A. annua) and has received great attention due to its antimalarial property. Research aimed at elucidating the biosynthetic route of the compound, has resulted in the isolation of four genes from the plant thus far (Bertea et al., 2006; Bouwmeester et al., 1999; Mercke et al., 2000; Rydén et al., 2010; Rydén et al., 2009; Wallaart et al., 1999; Zhang et al., 2008). The biosynthetic pathway of artemisinin has been clarified to the level of dihydroartemisinic acid (figure 1) (Bouwmeester et al., 1999; Mercke et al., 2000; Ro et al., 2006; Wallaart et al., 2001). The first committed precursor, amorpha-4,11-diene, is formed by cyclization of farnesyl diphosphate by amorphadiene synthase. Amorpha-4,11-diene is consecutively oxidized to artemisinic alcohol, artemisinic aldehyde and artemisinic acid by the cytochrome P450 Cyp71av1. Artemisinic aldehyde is also serving as a substrate for a carbon double bond reductase, Dbr2, which reduces the exocyclic carbon double bond, thereby forming dihydroartemisinic aldehyde. Dihydroartemisinic aldehyde is further oxidized by Cyp71av1 to dihydroartemisinic acid. Dihydroartemisinic acid, in contrast to artemisinic acid, is thought to be the immediate precursor to artemisinin. This is based on the finding that plants containing high levels of artemisinin also contain high levels of dihydroartemisinic acid, while plant with low levels of artemisinin display low levels of dihydroartemisinic acid and high levels of artemisinic acid. Currently, it is not clear whether the conversion of dihydroartemisinic acid to artemisinin occurs enzymatically or spontaneously in planta. Evidence supporting both claims has been presented. Recently, the conversion of dihydroartemisinic acid to artemisinin in dead leaves of A. annua has been reported (Lommen et al., 2007). However, the same laboratory also shows that dihydroartemisinic acid is converted to an unknown intermediate and

Chapter 6

166

PPO 111

FPP Amorpha-4,11-diene

1

15

10

6

14 13

11

AMDS

OH

H

Artemisinic alcohol

Cyp71av1

O

H


O

OH

Artemisinic acid

Cyp71av1

O

OH


Cyp71av1

OH

H


O

H


Red1

Artemisinin

Cyp71av1

Dbr2

Pox1Cyp71av1

Dbr1

O

OO

Arteannuin B

Artemisinin

Figure 1. Proposed biosynthesis of artemisinin. The identities of Dbr1 and Pox1 are not fully determined, but are supported by biochemical evidences reported in this chapter. Enzymatic conversion of arteannuin B has been reported (Dhingra et al., 2001).


167

thereafter into artemisinin in living as well as dead tissue in planta (Lommen et al., 2006). This unknown intermediate is postulated to be dihydroartemisinic acid hydrogen peroxide and can be formed non-enzymatically (Brown et al., 2004; Sy, 2002; Wallaart et al., 1999; Wallaart et al., 1999), as well as possibly enzymatically. The involvement of reactive oxygen species, such as hydrogen peroxide, in artemisinin production has been pointed out by adding salicylic acid to A. annua plants (Pu et al., 2009) and DMSO to A. annua shoot cultures (Mannan et al., 2010). There are also evidences that pin point artemisinic acid as the precursor of artemisinin. A link between artemisinic acid, arteannuin B and artemisinin has been established in hairy-root cultures (Weathers et al., 1997) and enzymatic conversion of arteannuin B has been presented (Dhingra et al., 2001), although the identity of the enzyme involved remains obscure.


Enzyme conversion assays

In order to evaluate a possible enzymatic conversion of either artemisinic acid or dihydroartemisinic acid to artemisinin, leaves and flowers of A. annua were collected and crude protein isolations were obtained from the fresh plant material. These protein solutions were tested with the substrates dihydroartemisinic aldehyde, artemisinic acid and dihydroartemisinic acid while adding the cofactor NADPH (figure 2) or hydrogen peroxide (figure 3 and figure 4).

As shown in figure 2, the production of dihydroartemisinic alcohol from the substrate dihydroartemisinic aldehyde by Red1 (figure 1) (Rydén et al., 2010) is dominant, as can be expected with NADPH added as cofactor, but there is also a significant production of artemisinic acid and traces of artemisinin. Artemisinic acid is

Chapter 6

168

DHAAOH

AAArtemisinin

Figure 2. GC chromatogram of assay with crude A. annua proteinisolation, dihydroartemisinic aldehyde and NADPH.

likely produced by a combination of NADP+ and an unknown carbon double bond reductase (Dbr1). NADP+ is produced by the action of Red1, which consumes NADPH and produces NADP+ and dihydroartemisinic alcohol. The substrate dihydroartemisinic aldehyde is oxidized to artemisinic aldehyde by Dbr1 while regenerating NADPH. Based on a published biochemical analysis of a known carbon double bond reductase from A. annua, Dbr2, it is likely that Dbr1 is different from Dbr2 (Zhang et al., 2008). The authors showed that Dbr2 converts artemisinic aldehyde to dihydroartemisinic aldehyde, but the reverse reaction was not reported. Artemisinic aldehyde, produced by Dbr1, is further oxidized to artemisinic acid by Cyp71av1. The steps between artemisinic acid and artemisinin are not clear.

The artemisinin intermediates artemisinic acid and dihydroartemisinic acid were exposed to crude A. annua protein isolations and hydrogen peroxide, to determine if the conversion to artemisinin is enzyme dependent or spontaneous (figure 3 and figure 4). Several cofactors such as KI and FeCl2 and other metal ions were investigated for a possible involvement in conversion of the acids to artemisinin, but neither enzymatic nor spontaneous


169

a

DHAA

*

b

DHAA

*

c

*

Chapter 6

170

d

DHAA*

e

DHAA

AA

*

f

DHAA

*


171

g

DHAA

*

Figure 3. Enzyme assays with crude protein isolations from A.annua flowers and leaves and dihydroartemisinic acid as substrate. AA – artemisinic acid, DHAA – dihydroartemisinic acid, * – unknown. The artemisinic acid in figure 3e is a small impurity of the substrate.

a. Crude protein extract with dihydroartemisinic acid as substrate and hydrogen peroxide.

b. Crude protein extract with dihydroartemisinic acid as substrate.

c. Crude protein extract with no substrates.d. Boiled crude protein extract with dihydroartemisinic

acid as substrate and hydrogen peroxide.e. Boiled crude protein extract with dihydroartemisinic

acid as substrate.f. Assay buffer with dihydroartemisinic acid and

hydrogen peroxide.g. Assay buffer with dihydroartemisinic acid.

activity could be detected using these conditions (data not shown). Figure 3 shows the results when crude protein isolations from A. annua leaves and flowers were tested with dihydroartemisinic acid and hydrogen peroxide as substrates. It is evident, that no enzymatic conversion occurs and also that there is no spontaneous conversion of dihydroartemisinic acid to artemisinin or a related intermediate. It is possible that other cofactors or cosubstrates will initiate a

Chapter 6

172

conversion of dihydroartemisinic acid. However, it is unlikely that dihydroartemisinic acid is converted to artemisinic acid and then to artemisinin (Bertea et al., 2005). The fate of dihydroartemisinic acid remains obscure.

On the other hand, the artemisinin intermediate artemisinic acid, is converted to arteannuin B (figure 4). This has been shown in A.annua hairy root cultures, but the responsible enzyme could not be identified (Arsenault et al., 2010). Arteannuin B has been shown to be enzymatically converted to artemisinin (Dhingra et al., 2001). Our finding point to a cytosolic protein, Pox1, which is dependent on hydrogen peroxide for production of arteannuin B from artemisinic acid. Pox1 is likely a protein that is involved in protection against reactive oxygen species and is probably overexpressed in stress situations in the plant.

a

AA

*

Arteannuin B


173

b

AA

*

c

AA

*

d

AA

*

Chapter 6

174

e

AA

*

f

AA

*

Figure 4. Enzyme assays with crude protein isolations from A.annua flowers and leaves and artemisinic acid as substrate. AA – artemisinic acid, * – unknown.

a. Crude protein extract with artemisinic acid as substrate and hydrogen peroxide.

b. Crude protein extract with artemisinic acid as substrate.

c. Boiled crude protein extract with artemisinic acid as substrate and hydrogen peroxide.

d. Boiled crude protein extract with artemisinic acid as substrate.

e. Assay buffer with artemisinic acid and hydrogen peroxide.

f. Assay buffer with artemisinic acid.


175

Protein pull down assay

In an effort to isolate Pox1 from A. annua, a protein pull down approach was taken (figure 5). This method is a one-step purification protocol and significantly speeds up the protein purification process compared to the traditional chemical biology approach (Kanoh et al., 2006; Kanoh et al., 2005). The method relies on the substrate molecules being randomly covalently attached to small beads. These beads are used to pull down proteins that interact with the substrate, thus facilitating purification and concentration of the target protein in one step. The sample is eluted from the beads using detergent and the sample is thereafter analyzed on an SDS-PAGE gel. The protein bands that appear from the eluted sample are cut out from the gel and analyzed with mass spectrometry and/or de novo sequencing. As our sample is originating from A. annua, simple mass spectrometry is not sufficient, as there is no protein fragment database to compare the results with. Therefore, it is necessary to perform de novo protein sequencing on the excized proteins.

Artemisinic acid was chosen as bate on the beads for a protein pull down assay based on the biochemical conversion data presented in this report (figure 4), which point to enzymatic conversion of artemisinic acid to arteannuin B, and literature supporting enzymatic conversion of arteannuin B to artemisinin (Dhingra et al., 2001). As shown in figure 5, three unique protein bands in the eluate can identified, whereof two are approximately the size of 50 kDa and a smaller protein of approximately 25 kDa. The sequencing of these bands is currently in process. After obtaing sequence fragments of the proteins and comparing them to the literature, the cDNA of the Pox1 candidate genes will be cloned from our trichome cDNA library (Bertea et al., 2006). These genes will thereafter be heterologously expressed in E. coli, purified and biochemically characterized. This work is in progress through a collaboration with Yokohama City University in Japan.

Chapter 6

176

1 32 4 5 6 7 1098

25 kDa

50 kDa75 kDa

Figure 5. 10% SDS-PAGE gel of protein pull down assay using artemisinic acid as capture molecule on the beads. Lanes were loaded with samples which were treated as described below.

1. Protein ladder marker. 2. Desalted crude protein isolation with bovine serum albumin. 3. Desalted crude protein isolation with bovine serum albumine,

supernatant from pre-clearing with control beads. 4. Desalted crude protein isolation with bovine serum albumine, wash

1, control beads. 5. Desalted crude protein isolation with bovine serum albumine, wash

2, control beads. 6. Desalted crude protein isolation with bovine serum albumine,

elution, control beads. 7. Desalted crude protein isolation with bovine serum albumine,

supernatant from pull down, beads with covalently attached artemisinic acid.

8. Desalted crude protein isolation with bovine serum albumine, wash 1, control beads, beads with covalently attached artemisinic acid.

9. Desalted crude protein isolation with bovine serum albumine, wash 2, control beads, beads with covalently attached artemisinic acid.

10. Desalted crude protein isolation with bovine serum albumine, elution, beads with covalently attached artemisinic acid.


177

If the protein fragments reveal interesiting candidates to the Pox1 enzyme, the chosen protein pull-down approach shows good potential for reducing the workload normally associated with chemical biology (Zhang et al., 2008).


Chemicals

Neat artemisinic acid, dihydroartemisinic acid and dihydroartemisinic aldehyde were synthesized as described by Bertea et al. (2005). Neat arteannuin B and artemisinin was isolated as described by Wallaart et al. (2000). Other chemicals were purchased from Oriental yeast Co. Tokyo, Japan or Sigma-Aldrich, USA. Beads used for the protein pull down assay were a gift from Hiroyuki Osada, RIKEN, Japan.

Plant material

Plants, originally identified by Wallaart et al., (1999) were grown for collection of flowers, leaf and root material using seeds obtained from University of Groningen. Seeds are deposited in the herbarium De Kruidhof in Buitenpost, the Netherlands, under registration number GR001. The A. annua line belongs to the high artemisinin chemotype and has the characteristic pattern of a high dihydroartemisinic acid to artemisinic acid ratio. Seeds were sown in potting compost and grown in climatrons with the conditions 21/18°C (16/8 h). Plants were watered as necessary without addition of fertilizer. When the plants started to flower, flowers and leaves were collected in four categories. Plant material was kept on ice during handling, then frozen in liquid nitrogen and stored at -80 °C.

Chapter 6

178

Crude protein isolation from A. annua

2 g Polyvinylpolypyrrolidone (PVPP) was soaked and gently mixed over night at 4°C in 50 ml buffer A (50 mM potassium phosphate buffer pH 7.5, 20% v/v glycerol, 2.5 mM EDTA pH 8.0, 0.2% BSA, pH 7.5). The slurry was pelleted and the buffer exchanged for 50 ml buffer B (50 mM potassium phosphate buffer pH 7.5, 20% v/v glycerol, 50 mM ascorbic acid, 50 mM natriummetabisulfiet, 5 mM DTT, 2.5 mM EDTA pH 8.0, 0.2% BSA, pH 7.5) supplemented with ½ a tablet of protease cocktail inhibitors (Roche). Plant material was chilled with liquid nitrogen and ground to a fine powder using a mortar and pestle. Thereafter 2 g powder was suspended in 15 ml buffer/PVPP slurry. The mixture was gently shaken for 10 min at 4°C and filtered through a fine nylon mesh to separate the fluid from PVPP and plant particles. The filtrate was centrifuged at 20.000 g for 20 min at 4°C and was thereafter de-salted on a PD-10 column with buffer B by following the instructions of the manufacturer (GE-Healthcare).

Enzyme conversion assays

Oxidoreductase assay

To 1 ml desalted crude protein isolation, 2.5 μl 10 mM dihydroartemisinic aldehyde was added as substrate and 1.7 mg NADPH as cofactor. The mixture was incubated 25°C for 4 hours. Thereafter the reaction was stopped on ice and 250 μl NaClaq was added. The mixture was repeatedly extracted with 1 ml hexane three times, and the organic phases were collected in one glass vial. The extract was carefully dried with a gentle nitrogen gas flow while cooled on ice. The solid remains in the vial were resuspended in 100 μl hexane and thereafter analyzed with GC-MS. Peaks were identified using authentic standards.


179

Peroxygenase assay

All boiled control samples were heated to 99°C for 10 min and cooled down on wet ice for 5 min before further processing. To 1 ml desalted crude protein isolation, 50 μl 2 mg/mL artemisinic acid or dihydroartemisinic acid was added per sample. To each sample, 11 μl 3% hydrogenperoxide was added and thereafter gently mixed followed by 30 min incubation at 25°C. The mixture was incubated for 4 hours and during this time, every 30 min 11 μl 3% hydrogenperoxide was added and the sample gently mixed. A total of 8 cycles with hydrogenperoxide replenishment was thus obtained. The reaction was stopped on ice and 250 μl NaClaq was added. The mixture was repeatedly extracted with 1 ml hexane three times, and the organic phases were collected in one glass vial. The extract was carefully dried with a gentle nitrogen gas flow while cooled on ice. The solid remains in the vial were resuspended in 100 μl hexane and thereafter analyzed with GC-MS. Peaks were identified using authentic standards.

Protein pull down assay

A 100 μl aliquot from the A. annua crude desalted protein isolation (described above) was taken as a control sample for analysis with SDS-PAGE. The beads used for the pull down assay were rinsed 3 times with ice-cold binding buffer (20 mM potassium phosphate buffer pH 7.5, 20% v/v glycerol, 150 mM KCl, final pH 7.5) to remove the storage buffer which contains sodium azide. 1 ml crude protein extract was pre-cleared by adding 50 μl rinsed control beads (100 μl 50% v/v slurry, pelleted and supernatant discarded). The mixture was incubated for 1 hour at 4°C with gentle shaking. Thereafter the mixture was pelleted and the pre-cleared supernatant transferred to 50 μl rinsed substrate beads. The mixture was incubated over night with gentle shaking for 6 hours at 4°C. Thereafter the sample was centrifuged and the supernatant saved in a microcentrifuge tube. The pellet was washed two times with 1 ml

Chapter 6

180

ice-cold binding buffer and to the final pellet was added 30 μl SDS-PAGE sample buffer (40% v/v glycerol, 8% SDS, 0.04% bromophenol blue, 5% beta-mercaptoethanol). The mixture was vigorously mixed and heated for 10 min at 99°C followed by a centrifugation in a bench top centrifuge at 12000 rpm for 10 min. The eluate was collected and analyzed on a 10% SDS-PAGE gel. 20 μl aliquots from the saved supernatants were analyzed on the same gel.

GC-MS

GC-MS analyses were performed on an Agilent 6890 series gas chromatograph system coupled to a JEOL JMS-SUN 200 mass selective detector in electron impact mode (70 eV) equipped with a guard column (1 m x 0.25 mm i.d., GL-Sciences) and a HP-5MS analytical column (30 m x 0.25 mm i.d., 0.25 μm film). The temperature program was set to an initial temperature of 60°C followed by a first temperature gradient of 3°C min-1 to 220°C and then a second gradient of 30°C min-1 to 300°C with an endpoint hold of 4 min. Helium flow was set to 0.7 ml min-1, detector temperature and injector temperature to 250°C using 500 scans min-1 with a scan range of 40-350 m/z.

References

Arsenault, P. R., Vail, D. R., Wobbe, K. K., & Weathers, P. J. (2010). Effect of Sugars on Artemisinin Production in Artemisia annua L.: Transcription and Metabolite Measurements. Molecules, 15(4), 2302-2318.



181




Dhingra, V., & Narasu, M. L. (2001). Purification and characterization of an enzyme involved in biochemical transformation of arteannuin B to artemisinin from Artemisia annua. Biochem Biophys Res Commun, 281(2), 558-561.

Kanoh, N., Asami, A., Kawatani, M., Honda, K., Kumashiro, S., Takayama, H., et al. (2006). Photo-Cross-Linked Small-Molecule Microarrays as Chemical Genomic Tools for Dissecting Protein-Ligand Interactions. Chemistry - An Asian Journal, 1(6), 789-797.

Kanoh, N., Honda, K., Simizu, S., Muroi, M., & Osada, H. (2005). Photo-Cross-Linked Small-Molecule Affinity Matrix for Facilitating Forward and Reverse Chemical Genetics13. Angewandte Chemie International Edition, 44(23), 3559-3562.

Lommen, W. J., Elzinga, S., Verstappen, F. W., & Bouwmeester, H. J. (2007). Artemisinin and sesquiterpene precursors in dead and green leaves of Artemisia annua L. crops. Planta Med, 73(10), 1133-1139.


Chapter 6

182

Mannan, A., Liu, C., Arsenault, P., Towler, M., Vail, D., Lorence, A., et al. (2010). DMSO triggers the generation of ROS leading to an increase in artemisinin and dihydroartemisinic acid in Artemisia annua shoot cultures. [10.1007/s00299-009-0807-y]. Plant Cell Reports, 29(2), 143-152.


Pu, G.-B., Ma, D.-M., Chen, J.-L., Ma, L.-Q., Wang, H., Li, G.-F., et al. (2009). Salicylic acid activates artemisinin biosynthesis in Artemisia annua L. [10.1007/s00299-009-0713-3]. PlantCell Reports, 28(7), 1127-1135.


Rydén, A.-M., Ruyter-Spira, C., Quax, W., Osada, H., Muranaka, T., Kayser, O., et al. (2010). The molecular cloning of dihydroartemisinic aldehyde reductase and its implication in artemisinin biosynthess in Artemisia annua. Planta Medica, Accepted.

Rydén, A.-M., Ruyter-Spira, C., Van den End, F., Verstappen, F., Quax, W. J., Osada, H., et al. (2009). Determination of oxygen as variable parameter for improved DHAAOH production.

Sy L.-K., B., G. D. . (2002). The mechanism of the spontaneous autoxidation of dihydroartemisinic acid. Tetrahedron, 58,897-908.



183




Weathers, P. J., Hemmavanh, D. D., Walcerz, D. B., Cheetham, R. D., & Smith, T. C. (1997). Interactive effects of nitrate and phosphate salts, sucrose, and inoculum culture age on growth and sesquiterpene production in Artemisia annua hairy root cultures. In Vitro Cellular & Developmental Biology - Plant 33(4), 306-312.


Chapter 7

Summary, general discussion and future perspectives

Chapter 7

186

Summary and persepectives

187

Summary, general discussion and future perspectives

Artemisia annua L. (A. Annua) is an annual herb growing in temperate and tropical regions of the world. It has the unique ability to produce an antimalarial compound, artemisinin, which has proved to be remarkably efficient in killing the malaria parasite while at the same time showing very minor side effects. The uniqueness of this compound and plant can be understood in the fact that A. annua is the only species capable of producing the artemisinin. From the moment when artemisinin was isolated and identified as the active ingredient against malaria in the 1970’s, the natural compound has been the focus for intense research in organic chemistry and biochemistry. However, the complete synthesis of artemisinin proved to be a daunting task and was deemed uneconomical.

The natural yield of artemisinin is very low; levels ranging around 0.2-0.8 % artemisinin (dry weight) are routinely reported. There are several breeding programs currently in progress which aim to increase the production of artemisinin in planta without the use of genetic engineering. This is a plausible way forward but this strategy is hampered by the same disadvantages as the original plant: unstable supply due to natural weather variations, poor final yield and devotion of fertile soil to non-food production. An alternative solution is the use of microorganisms as production platforms for artemisinin. This is achieved by identification of the genes in the biosynthetic pathway of artemisinin in A. annua and their implementation in a safe microorganism such as Saccharomyces cerevisiae (S. cerevisiae). This allows for a stable, high, economical and controlled production of the drug. Furthermore, the knowledge of the genes involved in the pathway opens up for potential genetic improvement of the plant in terms of artemisinin yield should such an aim ever be accepted and desirable.

In this thesis, the biosynthetic genes of artemisinin in A. annua are identified and their potential in the heterologous expression platform

Chapter 7

188

S. cerevisiae is investigated. Chapter 1 and 2 provides background information of the history and chemistry of artemisinin. The suitability of S. cerevisiae as a production platform for artemisinin is discussed as well and examples are given for the implementation of synthetic biology.

Isolation of a broad substrate reductase from A. annua withimplications in the artemisinin biosynthetic pathway

In chapter 3 and chapter 4, data are presented relating to a new reductase isolated from the plant A. annua. In an effort to isolate genes involved in the biosynthetic pathway of artemisinin an EST library was sequenced and analyzed against other EST libraries and the published database collected at blastX (NCBI). A first analysis using blastX revealed 19 candidates out of 2180 EST sequences which possessed characteristic amino acid combinations of the reductase family. However, the enzymatic identities of the reductases remained largely obscure as the experimentally validated knowledge accumulated in the public database is very limited for plant reductases. To create a ranking list with the most promising candidates at the top, comparisons with EST libraries from tomato, potato, rice, Arabidopsis thaliana (A. thaliana), lettuce and sunflower EST libraries were pursued. To distinguish between ubiquitous reductases and reductases likely to be relevant in terpenoid biosynthesis, the following strategy was adopted: Reductases with a high similarity with lettuce and sunflower but significantly lower similarity with the non-Asteraceae outgroups potato (Solanaceae), tomato (Solanaceae), rice (Poaceae) and A.thaliana (Brassicaceae) were considered to be promising candidates in the terpenoid biosynthesis. Based on the bioinformatics analysis a single candidate, red1, was picked for further experimental studies. Several substrates, all members of the chemical class terpenoids, were tested for conversion using the cofactor NADP(H). The results presented in chapter 3 clearly show that Red1 is a broad substrate oxidoreductase belonging to the short chain dehydrogenase/reductase


189

(SDR) protein family. It is particularly efficiently reducing (-)-menthone to (+)-neomenthol and can oxidize (+)-neomenthol to (-)-menthol via the intermediate (-)-menthone. This at efficiency rates rivalling those of the industrially important plant Mentha x piperitea.The menthone:neomenthol reductase of Mentha x piperitea attain a kcat/Km of 89 (M-1, s-1) which is the previously highest value recorded for a plant reductase whereas Red1 display a kcat/Km of 83620 (M-1, s-

1), a 940 fold difference in efficiency. Other substrates such as dihydrocarvone and perilla aldehyde are converted by Red1 as well. All these compounds belong to the terpenoid subgroup monoterpenoids (C5). However, the terpenoids in the artemisinin biosynthetic pathway belong to the subgroup sesquiterpenoids (C15). A wider range of sesquiterpenoids were tested for conversion by Red1 but were consistently refused. As perilla aldehyde has a chemical similarity to artemisinic aldehyde and dihydroartemisinic aldehyde, these substrates were tested for conversion by Red1 (chapter 4). Red1 is a potent reducer of dihydroartemisinic aldehyde and will produce dihydroartemisinic alcohol while refusing the alcohol as a substrate. Red1 produce a kcat/Km value of 4119 (M-1, s-1)to be compared with another devoted reductase in the pathway, the carbon double bond reductase Dbr2 which show a kcat/Km value of 0.14 (M-1, s-1).

The impact of Red1 in the biosynthetic pathway of artemisinin can be better understood if one thinks in networks rather than simple straight down stream pathways. The effect of Red1 is to remove a valuable intermediate in the pathway, the dihydroartemisinic aldehyde. Currently there is only one reported enzyme which is able, albeit at low levels, to convert the alcohol back into the aldehyde so that it in its turn can be used for the production of artemisinin (chapter 5). We can speculate about the role of Red1 in the pathway; it may serve as a regulator of the flux or it might simply be one of many tasks performed in planta and the reduction of the aldehyde may simply be a, for our purpose, misfortunate side effect. It is, however, likely that the expression of Red1 on the artemisinin

Chapter 7

190

production in A. annua is detrimental to the artemisinin yield. Currently a project aiming to develop a transformation system for A. annua is in progress and it is estimated that in the spring 2011 this system can be used to transform A. annua with a siRNA construct which will downregulate the expression of red1. The effect thereof is expected to lead to increased levels of artemisinin in the plant.

Investigating Saccharomyces cerevisiae as a drug production platform

Plants aside, the knowledge of Red1 and its function reminds us that it is important to investigate potential host platforms for their metabolic reaction on heterologous compounds. It may well be that the host already contains native genes that will perform functions in the heterologous pathway. If such genes can be identified, fewer foreign genes need to be introduced into the host and hence the stress on the organism is limited. To maximize the yield for the desired product it is important to reduce the stress on the organism. S.cerevisiae is a convenient host which is well characterized and which also produces the necessary precursors for artemisinin production. One way to adapt it to produce artemisinin is simply to transfer all the biosynthetic genes of the pathway from A. annua to the yeast. This, however, is putting a considerable stress on the organism which aims by various means to limit the production of drug intermediates either by excluding the genes or by over expressing stress response genes which results in an efflux of drug intermediates from the host. One strategy to lessen the stress response caused by expression of foreign genes is to employ the function of endogenous genes and to use them instead of foreign genes for production of the heterologous intermediates. In chapter 5 is described how oxygen limitation and the choice of carbon source can change the genetic expression pattern of S. cerevisiae in such a way that it can convert an early intermediate into an interesting reduced intermediate which is dedicated to the production of artemisinin.


191

Future perspectives

A project aimed at adapting a protein fishing method has been developed in collaboration with RIKEN and Yokohama City University (chapter 6). The goal is to speed up the protein discovery procedure by using a reverse genetics approach. By coating beads with the substrate of the unknown protein and applying a crude extract of plant proteins it is possible to purify only the proteins that react with the substrate coated beads. Artemisinic acid is the last proven biosynthetic step in the artemisinin pathway. A pilot project using artemisinic acid coated on the beads and crude protein extract from A. annua produced two distinct protein bands on an SDS-PAGE gel. De novo sequencing of these proteins is currently in progress and the resulting amino acid sequence will be used to pick up candidates from the EST library. In parallel, biochemical work was performed to establish the next biosynthetic product after artemisinic acid and a biochemical test system was designed for later confirmation of enzymatic activity of the expressed cloned candidate genes.

Furthermore, the potential of different organelles in S. cerevisiae was delineated. Currently, synthetic biology efforts are focusing on manipulation in the cytoplasm of the host organism due to the easiness of this procedure. This ignores the potential of the mitochondria and peroxisomes as production places of the artemisinin intermediates. Mitochondria have a well documented large pool of the immediate precursor to the first artemisinin intermediate amorpha-4,11-diene. The prevalence of the necessary precursor in peroxisomes is uncertain although in human cells this precursor does exist. Constructs expressing amorphadiene synthase were designed in such a way as to target the synthase to either the cytoplasm, the mitochondria or the peroxisome. It would be interesting to see if both organelles can significantly contribute to the artemisinin intermediate production. It would be a benefit if both the mitochondria and the peroxisome could be used as they complement

Chapter 7

192

each other in terms of carbon utility and hence allow for a maximized utilization of input material as the yeast would be able to produce the artemisinin intermediate during the entire batch cycle.

The suitability of S. cerevisiae was confirmed as a first choice for pilot studies. Initially, studies on Xanthophyllomyces dendrorhous and its ability to express amorphadiene synthase and produce amorpha-4,11-diene, the first committed compound in the artemisinin biosynthetic pathway, were performed. As the genetic tools are very limited for this organism, the gene was incorporated into redundant rRNA. Several copies of the genes were shown to be incorporated into the host but the expression of the enzyme proved irregular. It is possible that this is due to silencing of the redundant rRNA and possible the host can sense the foreign DNA and class it as a threat and deliberately silence that area. Currently there is no way of controlling which DNA stretch in rRNA will be active or inactive. At this moment there is a project in progress aiming to develop new genetic tools for X. dendrorhous and it is our hope that this organism will prove a better host than S. cerevisiae in terms of artemisinin intermediate yields.

Chapter 8

Nederlandse samenvatting

Chapter 8

194


195

Samenvatting

Introductie

Artemisia annua (A. annua), de eenjarige alsem, is een plant die wereldwijd voorkomt in gematigde en tropische gebieden. A. annua behoort tot de plantenfamilie Asteraceae (samengesteldbloemigen, composieten). De plant produceert artemisinine, een antimalariamiddel dat malariaparasieten efficiënt doodt, zonder veel bijwerkingen voor de patiënt. Van nature is het gehalte aan artemisinine in de plant laag en ligt tussen 0,2 en 0,8%, berekend op drooggewicht. Er lopen verschillende veredelingsprogramma’s die erop gericht zijn de productie van artemisinine in de plant te verhogen, zonder daarbij gebruik te maken van genetische modificatie. Deze aanpak kan een stap in de goede richting zijn, maar wordt beperkt door weersinvloeden die de groei van de plant en dus de uiteindelijke opbrengst negatief kunnen beïnvloeden. Daarnaast moet vruchtbare grond worden gebruikt voor het kweken van gewassen die niet als voedingsmiddel dienen. Een alternatieve aanpak is het gebruik van micro-organismen om artemisinine te produceren. Het micro-organisme is dan een zogenaamd heteroloog expressiesysteem of heterologe gastheer. Dit is alleen mogelijk als we de genen die betrokken zijn bij de biosynthese van arteminine in A. annua kennen, waarna we deze overzetten in een veilig micro-organisme, zoals de gist Saccharomyces cerevisiae (S. cerevisiae).Hiermee kunnen we het geneesmiddel in principe op een stabiele manier, gecontroleerd en economisch in grotere hoeveelheden produceren. Daarnaast opent kennis van de genen die bij de biosynthese betrokken zijn de mogelijkheid om de plant genetisch te verbeteren om zo tot een hogere productie van artemisinine te komen.

In dit proefschrift beschrijven we de identificatie van genen in A.annua die betrokken zijn bij de biosynthese van artemisinine, en de mogelijkheid om deze genen tot heterologe expressie te brengen in S.

Chapter 8

196

cerevisiae. Hoofdstuk 1 en 2 geven achtergrondinformatie over de geschiedenis en de chemie van het het terpenoïd artemisine. We bespreken de toepassing van S. cerevisiae als heteroloog productie-organisme voor artemisinine tegen de achtergrond van andere voorbeelden uit de synthetische biologie.

Isolatie van een reductase uit A. annua met brede substraatspecificiteit en de rol ervan in de biosynthese van artemisinine

In hoofdstuk 3 en 4 beschrijven we de isolatie van een nieuw reductase uit A. annua. We hebben voor ons werk aan de isolatie van genen uit de biosyntheseroute van artemisinine een EST-bibliotheek gemaakt en deze vergeleken met andere EST-bibliotheken en databases die op blastX (NCBI) zijn gepubliceerd. De eerste analyse met behulp van blastX leverde 19 kandidaten op uit een totaal van 2180 EST-sequenties, met aminozuurcombinaties die kenmerkend zijn voor de reductase enzymfamilie. Omdat de informatie in de databases erg beperkt is voor reductases uit planten, bleef de identiteit van onze enzymen grotendeels onduidelijk. Om tot een lijstje te komen met de meest veelbelovende kandidaten, vergeleken we onze gegevens met EST-bibliotheken van tomaat, aardappel, rijst, Arabidopsis thaliana (A. thaliana; zandraket), sla en zonnebloem. We wilden onderscheid maken tussen algemeen voorkomende reductases en reductases die waarschijnlijk relevant zijn in de biosynthese van terpenoïden. Reductases die grote overeenkomst vertoonden met sla en zonnebloem (die net als A. annua behorenen tot de samangesteldbloemigen), maar minder met de de niet-Asteraceae aardappel (Solanaceae, nachtschadefamilie), tomaat (Solanaceae), rijst (Poaceae, grassen) en A. thaliana (Brassicaceae, kruisbloemigen) beschouwden wij als interessante kandidaten. Gebaseerd op een bioinformatica-analyse van één van de kandidaten, selecteerden wij het gen red1 voor verder onderzoek. We testten diverse substraten, allemaal behorend tot de chemische klasse van terpenoïden, voor bioconversie (omzetting door het enzym) met


197

NADP(H) als cofactor. De resultaten in hoofdstuk 3 laten zien dat Red1 een oxidoreductase is met brede substraatspecificiteit, die behoort tot de korte-keten dehydrogenase/reductase (SDR) eiwitfamilie. Dit enzym kan op efficiënte wijze (-)-menthon reduceren tot (+)-neomenthol, en (+)-neomenthol oxideren tot (-)-menthol, met (-)-menthon als tussenproduct. De efficiëntie waarmee deze omzettingen plaatsvonden waren vergelijkbaar met die van de industrieel belangrijke plant Mentha x piperita (pepermunt). Het menthon:neomentholreductase van Mentha x piperita had een kcat/Kmvan 89 (M-1,s-1), tot dan toe de hoogste waarde bepaald voor een plantenreductase. Red1 bezat een kcat/Km van 83620 (M-1,s-1), een 940 keer hogere waarde. Andere substraten, zoals dihydrocarvon en perilaldehyde konden ook door Red1 worden omgezet. Al deze verbindingen behoren tot de groep van sesquiterpenoïden (C15). We testten vervolgens een bredere range van sesquiterpenoïden, maar deze werden niet omgezet door het enzym. Aangezien perilaldehyde chemisch gezien lijkt op artemisinine aldehyde en dihydroartemisinine aldehyde, testten we deze substraten ook op omzetting door Red1 (hoofdstuk 4). Red1 bleek een sterke reductor van dihydroartemisinine aldehyde, waarbij dihydroartemisinine alcohol werd gevormd, maar kon het alcohol niet als substraat gebruiken. Red1 voerde de reductie uit met een kcat/Km van 4119 (M-

1,s-1) en was daarmee vergelijkbaar met een ander reductase in de biosyntheseroute, het Dbr2 dat een dubbele koolstofbinding kan reduceren, met een kcat/Km van 0,14 (M-1,s-1).

Het belang van Red1 in de vorming van artemisinine in de plant kunnen we beter begrijpen als we denken in termen van netwerken in plaats van simpele, rechttoe rechtaan biosynthesewegen. Red1 verwijdert dihydroartemisinine aldehyde, een belangrijk intermediair in de biosyntheseroute. Op dit moment is slechts één enzym bekend dat in staat is om, in kleine hoeveelheden, het alcohol terug om te zetten in het aldehyde en daarmee beschikbaar te maken voor de productie van artemisinine (hoofdstuk 5). We kunnen speculeren over de rol die Red1 speelt in de biosyntheseroute. Het kan van de

Chapter 8

198

stroom van intermediairen in de biosynthese reguleren, maar het kan ook een heel andere taak hebben in de plant, waarbij de reductie van het aldehyde in ons geval een ongewenst bij-effect is. Wel is het waarschijnlijk dat de expressie van Red1 in A. annua nadelige gevolgen heeft voor de productie van artemisinine. Momenteel loopt er een project dat la doel heeft een transformatiesysteem voor A. annua te ontwikkelen. Naar verwachting kan het systeem in het voorjaar van 2011 worden gebruikt om A. annua te transformeren met een siRNA-construct, dat de expressie van red1 moet terugbrengen (downreguleren). Het effect hiervan moeten hogere concentraties artemisinine in de plant zijn.

Onderzoek naar Saccharomyces cerevisiae als productie-organisme voor artemisinine

De kennis over Red1 en de de functie van dit enzym hebben ons aangezet om de mogelijkheid van een heteroloog expressie-organisme voor artemisinine te onderzoeken. Bij heterologe expressie wordt het gen ingebouwd in een gastorganisme, waarin het niet van nature voorkomt. Het kan ook zo zijn dat het gastorganisme genen bevat die reacties uit de voor dat organisme vreemde, heterologe route kunnen uitvoeren. Als we in staat zijn om deze genen te identificeren, hoeven we minder vreemde genen in het gastorganisme in te bouwen en oefenen we hierop minder stress uit. Dit laatste is van belang om de opbrengst van het gewenste product later te vergroten. S. cerevisiae, is een veel gebruikte en goed gekarakteriseerde gastheer, die ook zelf de nodige precursors voor artemisinine maakt. Een manier om het organisme aan te zetten tot artemisinineproductie is simpelweg door het overbrengen van alle genen uit de plant die bij de biosynthese betrokken zijn. Dit veroorzaakt echter grote stress en de gist zal proberen de productie van intermediairen uit de vreemde biosynthese te beperken, hetzij door selectief genen uit te schakelen, hetzij door het tot overexpressie brengen van ‘stress response’genen waardoor intermediairen uit het gastorganisme worden geëlimineerd. Een


199

manier om de stress door vreemde genen te beperken, is gebruikmaken van genen van het gastorganisme zelf en deze de heterologe intermediairen laten aanmaken. In hoofdstuk 5 beschrijven we hoe een beperking van zuurstof en de keuze van een koolstofbron het genetische expressiepatroon van S. cerevisiae zo kunnen veranderen, dat het een vroege intermediair uit de biosynthesroute van artemisinine kan reduceren in een product dat relevant is voor de vorming van het eindproduct.

Betrokkenheid van peroxidase in de artemisinineproductie

In samenwerking met RIKEN en de Oklahoma City University werd te ‘protein fishing’ als methode gebruikt om enzymen te isoleren (hoofdstuk 6). Het doel was om de procedure van het vinden van eiwitten te versnellen met behulp van ‘reverse genetics’. Door bolletjes met substraat voor een onbekend eiwit te coaten en gebruik te maken van ruwe eiwitextracten van planten, is het mogelijk alleen die eiwitten te isoleren die met het substraat op de bolletjes reageren. Artemisininezuur is de laatste precursor in de biosynthese van artemisinine. Met de hierboven geschetste aanpak verkregen wij in een pilotexperiment met een eiwitextract van A. annua twee duidelijke banden op een SDS-PAGE gel. Momenteel werken we aan het sequencen van deze eiwitten, waarna we de aminozuurvolgorde zullen gebruiken om kandidaten op te pikken uit de EST-bibliotheek. Tegelijkertijd hebben we biochemisch werk gedaan gericht op het vaststellen van de volgende biosynthesestap, na de vorming van artemisininezuur. We ontwierpen een biochemisch testsysteem om later enzymactiviteit te kunnen bevestigen in gekloonde en tot expressie gebrachte kandidaatgenen.

Acknowledgements

Acknowledgements

202

Acknowledgements

203

The work presented in this thesis, and other projects which could not be fitted into these pages, owe their existence to many people. As is true for all PhD students, their training, development and success is dependent on support and advice from promotores, co-workers and colleagues. During this time I have been privileged to be included in a network of scientists and friends within Groningen and Wageningen, who also opened up the world to me. I am most grateful to all these people and would like to convey my sincere appreciation. Thanks to all these people, I look forward to explore, manage and produce a scientific heritage together with co-workers and students, for future academic generations.

What is the tree without its roots? I thank Prof. dr. Oliver Kayser and Prof. dr. Wim J. Quax for welcoming me in the Department of Pharmaceutical Biology, for their invaluable guidance and for their continuing support. I am grateful to my promotores, who supported and encouraged me in my desire to try new research strategies and create international collaborations. I am confident that I am heading out well prepared for the academical battle field and thank you from the bottom of my heart.

I thank Prof. dr. Harro Bouwmeester at Wageningen University for his collaboration and guidance in the artemisinin project. You welcomed me into your laboratory early on during my PhD study, and I recurred into your laboratory so often that it felt like my second work space apart from Groningen University. While you were still appointed at Plant Research International, I shared many Friday afternoons in relaxed company with our colleagues, while getting updated on foreign specialities. I arrived confused and gradually learned how to do research and think as a researcher. I am grateful for your commitment and optimism.

During my time in Wageningen, I met many people to whom I am grateful for their friendship and scientific guidance. I would especially like to thank Carolien Ruyter-Spira, who opened her home

Acknowledgements

204

and let me share the family Dutch family life. I am grateful for all the discussions we had, scientific as well as life oriented. Thank you for teaching me so much about science and life. I thank Fracel Verstappen, without whose expertise in analytical biochemistry I would have fallen short. You sacrificed many off your “free” days for science, and this engagement is a memory that will continue to inspire me. I thank Jules Beekwilder for his many advices on protein isolation and enzyme kinetics and for showing me that there is a time to be relaxed in science and other times to be meticulous. I thank Fred van den Eend who taught me the secrets about fermentation, Teun van Herpen for friendly as well as scientific discussions, Ralph Litjens for teaching me molecular biology techniques, Maarten Jongsma for scientific discussions, Geert Stoopen for enzyme isolation protocols and Margriet Roelse for SDS-PAGE instructions. I thank all my other wonderful colleagues at the Plant Research International who all contributed to the seamless transitions between Groningen University and Plant Research International.

To my colleagues at Groningen University, my presence at the institute must have been ghost like. I know you sometimes wondered whether I am still with you or already had moved on to other positions. Still, every time I returned, you all welcomed me back as if I had never been gone. I thank you all for the friendly atmosphere and the mutual encouragement. I thank Gerrit Poelarends for his scientific comments and advices, Mattijs Julsing for showing me his preferred method of protein blotting and for your discussions and friendship, Remco Muntedam for interesting discourses and for sharing the hustle of co-authorship, Elena Melillo for your friendship, generosity and scientific collaboration, Rita Setroikromo for invaluable assistance, your kindness and for opening your home to me, Ronald van Merkerk for keeping the lab running, Janita Zwinderman and Yvonne Vergnes-Mettes for keeping track of me and easing the bureaucracy. I thank Carlos R. Rodrigues dos Reis, Vinod Puthan Veetil, Hans Raj, Bert Jan Baas, Anna Wasiel, Torsten Flemming, Oktavia Hendrawati, Remco Muntendam, Luis F. da

Acknowledgements

205

Silva Godinho, Mariette Heins, Gudrun Koch, Aart van Assen, Ilse de Haan, Mariana Wahjudi, Magda Czepnik, Pol Nadal Jimenez, Evelina Papaioannou, Muhamad Insanu and Gany Herianto for making our department such a special place. I thank my students, through whom I learned how to be a better teacher. I wish you all success and fortune in your future.

I thank the reading committee: for their revision of this manuscript.

I thank Oktavia Hendrawati and Magda Czepnik for shouldering the role as paranimfen during my defence.

I thank Prof. dr. Hiroyuki Osada at the Chemical Biology Department in RIKEN in Tokyo, for his collaboration and for lending me his post doctoral fellows and senior staff.

I thank Prof. dr. Toshiya Muranaka at the Kihara Institute for Biological Research at Yokohama City University in Yokohama, for his collaboration and for the assistance of his senior researchers and post doctoral fellows.

I thank Dr. Herman J. Woerdenbag for his Dutch translation of the English summary.

I give special thanks to the board of Gålöstiftelsen, who decided to support my research with a stipend. This stipend made it possible for me to visit the lab of Cold Spring Harbor in New York, USA, in order to participate in molecular biology courses. I am happy to be able to say that the gathered knowledge turned out to be crucial for my research project and that it will continue to beneficial for years to come.

Once again, I thank all my international and local colleagues for their help which emanated in this thesis and many collaboration projects, some of which are still active.

Acknowledgements

206

I thank my family for their unconditional and full support.

I give a special warm thought to Khan who supported me in finalizing this thesis.

Thank you all, it has been an adventurous journey.


Publication list

Publication list

208

Publication list

209

Peer reviewed publications

1. Rydén A-M, Ruyter-Spira C, Quax W. J., Osada H, Muranaka T, Kayser O, Bouwmeester H. J., Molecular cloning of dihydroartemisinic aldehyde reductase and its implication in artemisinin biosynthesis. In press. Planta Medica, 2010.

2. Rydén A-M, Ruyter-Spira C, Litjens R, Takahashi S, Quax W. J., Osada H, Bouwmeester H. J., Kayser O, Molecular cloning and characterization of a broad substrate terpenoid oxidoreductase from Artemisia annua.In press. Plant and Cell Physiology, 2010.

3. Rydén A-M, Ruyter-Spira C, Van den End F, Verstappen F, Quax W. J., Bouwmeester H. J., Kayser O, Oxygen dependent biosynthesis of artemisinin precursors in yeast. Submitted, Journal of biotechnology, 2010.

Reviews

4. Muntendam, R, Melillo, E, Rydén A-M, Kayser O, Perspectives and Limits of engineering the isoprene metabolism in heterologous hosts. In press, Applied microbiology and biotechnology, 2009; vol. 84, No. 6, p. 1003-1019.

5. Rydén A-M, Kayser, O, Sesquiterpene trioxane lactones – chemistry and biological activities. Topics in heterocyclic chemistry, Springer, Vol. 9, pp. 1-28, 2007.

Publication list

210

Dissertation

6. Rydén A-M Role of peroxygenase in artemisinin production. Doctoral thesis, University of Groningen, 2010.

Poster presentations

2009 Chemotype shifting in yeast: Minimizing heterologous genetic load and maximizing heterologous product yield. TERPNET, Tokyo, Japan.

2009 Cloning and characterization of an Artemisia annua terpenoid reductase implicated in the biosynthetic pathway of the antimalarial drug artemisinin. TERPNET, Tokyo, Japan.

2007 Cloning and functional elucidation of cytochrome P450s from Artemisia annua. TERPNET, Strasbourg, Frankrike.

2006 Biosynthesis of artemisinin in Artemisia annua. GUIDE, Groningen, Holland.

211

Curriculum vitae

Curriculum Vitae

212

Curriculum Vitae

213

Professional experience

National Cancer Center Tokyo, Japan 2010-2012

Japan Society for the Promotion of Science postdoctoral fellow. Elucidating the regulatory mechanism and therapeutic prospects of Parp protein in carcinogenesis.

University of Groningen Groningen, the Netherlands 2005-2010

Doctoral studies in pharmaceutical biology. Product development of anti-malarial drug. Designed and conducted experiments. Published in international scientific journals at international conferences.

Yokohama City UniversityYokohama, Japan 2009

Established collaboration with Yokohama City University as invitedresearcher. Studies of Japanese organizations models and initiation of research collaboration.

RIKENWako-shi, Japan 2008-2009

Established collaboration with the research institute RIKEN as invitedresearcher. Studies of protein isolation techniques and initiation of research collaboration.

Plant Research InternationalWageningen, the Netherlands 2005- 2008

Established collaboration with the research institute Plant Research International as a recurring invited researcher. Product development of anti-malarial drug. Initiation of research collaboration.

Curriculum Vitae

214

Education

Cold Spring Harbor LaboratoryNew York, USA2007

Biotechnological courses during 3 months in yeast genetics and genomics and molecular techniques in plant science.

Technical University BerlinBerlin, Germany 2004- 2005

Master thesis. Title: Virulence genes of Entercoccus faecalis.

Technical University BerlinBerlin, Germany 2003- 2004

ERASMUS-student at Technical University of Berlin during final theoretical year of M.Sc. Free moving student at Humboldt University Berlin and Free University Berlin.

Royal Institute of TechnologyStockholm, Sweden 2000-2006

Master of Science in Biotechnology.Broad education in life science with strong mathematical and chemical influence.

Uppsala University Uppsala, Sweden Summer 2000

Research school. Thesis: Impact of pollutants on marine microbiological ecology.

Danderyds gymnasium Stockholm, Sweden 1998-2000

Senior high school studies in naturalsciences.

Curriculum Vitae

215

Stipends and awards

2010-2012 Postdoctoral fellowship, Japan Society for the Promotion of Science, Japan.

2005-2010 Ubbo Emmius stipend, University of Groningen, the Netherlands.

2007 Stipend from GUIDE, University of Groningen, the Netherlands.

2006 Stipend from Sixten Gemzéus stiftelse, Gålöstiftelsen, Sweden.

2003 ERASMUS stipend, European Union.

2000 Stipend for research school in Uppsala, Sweden.

2000 Stipend for excellence in biology, mathematics, chemistry and physics. Principal Ernst Helin’s stipend, Danderyds gymnasium, Sweden.

1994-2000 Chemistry stipend from “Berzeliusdagarna”. Awards for excellence in design, art and technology. Alternative Nobel prize in social science. Several awards for theoretical studies.

Curriculum Vitae

216

University of Groningen Identification, characterisation ... · treatments for uncomplicated...

Documents

Transcript of University of Groningen Identification, characterisation ... · treatments for uncomplicated...