Regioselective biooxidation of (+)-valencene by recombinant E. coli expressing CYP109B1 from Bacillus subtilis in a two-liquid-phase system

Background  

(+)-Nootkatone (4) is a high added-value compound found in grapefruit juice. Allylic oxidation of the sesquiterpene (+)-valencene (1) provides an attractive route to this sought-after flavoring. So far, chemical methods to produce (+)-nootkatone (4) from (+)-valencene (1) involve unsafe toxic compounds, whereas several biotechnological approaches applied yield large amounts of undesirable byproducts. In the present work 125 cytochrome P450 enzymes from bacteria were tested for regioselective oxidation of (+)-valencene (1) at allylic C2-position to produce (+)-nootkatone (4) via cis- (2) or trans-nootkatol (3). The P450 activity was supported by the co-expression of putidaredoxin reductase (PdR) and putidaredoxin (Pdx) from Pseudomonas putida in Escherichia coli.     

Amorpha-4,11-diene synthase: cloning and functional expression of a key enzyme in the biosynthetic pathway of the novel antimalarial drug artemisinin

The sesquiterpenoid artemisinin, isolated from the plant Artemisia annua L., and its semi-synthetic derivatives are a new and very effective group of antimalarial drugs. A branch point in the biosynthesis of this compound is the cyclisation of the ubiquitous precursor farnesyl diphosphate into the first specific precursor of artemisinin, namely amorpha-4,11-diene. Here we describe the isolation of a cDNA clone encoding amorpha-4,11-diene synthase. The deduced amino acid sequence exhibits the highest identity (50%) with a putative sesquiterpene cyclase of A. annua. When expressed in Escherichia coli, the recombinant enzyme catalyses the formation of amorpha-4,11-diene from farnesyl diphosphate. Introduction of the gene into tobacco (Nicotiana tabacum L.) resulted in the expression of an active enzyme and the accumulation of amorpha-4,11-diene ranging from 0.2 to 1.7 ng per g fresh weight. Received: 8 June 2000 / Accepted: 21 August 2000     

Production of the Artemisinin Precursor Amorpha-4,11-diene by Engineered Saccharomyces cerevisiae

The gene encoding for amorpha-4,11-diene synthase from Artemisia annua was transformed into yeast Saccharomyces cerevisiae in two fundamentally different ways. First, the gene was subcloned into the galactose-inducible, high-copy number yeast expression vector pYeDP60 and used to transform the Saccharomyces cerevisiae strain CEN·PK113-5D. Secondly, amorpha-4,11-diene synthase gene, regulated by the same promoter, was introduced into the yeast genome by homologous recombination. In protein extracts from galactose-induced yeast cells, a higher activity was observed for yeast expressing the enzyme from the plasmid. The genome-transformed yeast grows at the same rate as wild-type yeast while plasmid-carrying yeast grows somewhat slower than the wild-type yeast. The plasmid and genome-transformed yeasts produced 600 and 100 μg/l of the artemisinin precursor amorpha-4,11-diene, respectively, during 16-days’ batch cultivation. Revisions requested 14 November 2005; Revisions received 17 January 2006     

Production of the sesquiterpenoid (+)-nootkatone by metabolic engineering of Pichia pastoris

The sesquiterpenoid (+)-nootkatone is a highly demanded and highly valued aroma compound naturally found in grapefruit, pummelo or Nootka cypress tree. Extraction of (+)-nootkatone from plant material or its production by chemical synthesis suffers from low yields and the use of environmentally harmful methods, respectively. Lately, major attention has been paid to biotechnological approaches, using cell extracts or whole-cell systems for the production of (+)-nootkatone. In our study, the yeast Pichia pastoris initially was applied as whole-cell biocatalyst for the production of (+)-nootkatone from (+)-valencene, the abundant aroma compound of oranges. Therefore, we generated a strain co-expressing the premnaspirodiene oxygenase of Hyoscyamus muticus (HPO) and the Arabidopsis thaliana cytochrome P450 reductase (CPR) that hydroxylated extracellularly added (+)-valencene. Intracellular production of (+)-valencene by co-expression of valencene synthase from Callitropsis nootkatensis resolved the phase-transfer issues of (+)-valencene. Bi-phasic cultivations of P. pastoris resulted in the production of trans-nootkatol, which was oxidized to (+)-nootkatone by an intrinsic P. pastoris activity. Additional overexpression of a P. pastoris alcohol dehydrogenase and truncated hydroxy-methylglutaryl-CoA reductase (tHmg1p) significantly enhanced the (+)-nootkatone yield to 208 mg L⁻¹ cell culture in bioreactor cultivations. Thus, metabolically engineered yeast P. pastoris represents a valuable, whole-cell system for high-level production of (+)-nootkatone from simple carbon sources.     

Biosynthesis of costunolide,dihydrocostunolide, and leucodin. Demonstration of cytochrome p450-catalyzed formation of the lactone ring present in sesquiterpene lactones of chicory

Chicory (Cichorium intybus) is known to contain guaianolides, eudesmanolides, and germacranolides. These sesquiterpene lactones are postulated to originate from a common germacranolide, namely (+)-costunolide. Whereas a pathway for the formation of germacra-1(10),4,11(13)-trien-12-oic acid from farnesyl diphosphate had previously been established, we now report the isolation of an enzyme activity from chicory roots that converts the germacrene acid into (+)-costunolide. This (+)-costunolide synthase catalyzes the last step in the formation of the lactone ring present in sesquiterpene lactones and is dependent on NADPH and molecular oxygen. Incubation of the germacrene acid in the presence of 18O2 resulted in the incorporation of one atom of 18O into (+)-costunolide. The label was situated at the ring oxygen atom. Hence, formation of the lactone ring most likely occurs via C6-hydroxylation of the germacrene acid and subsequent attack of this hydroxyl group at the C12-atom of the carboxyl group. Blue light-reversible CO inhibition and experiments with cytochrome P450 inhibitors demonstrated that the (+)-costunolide synthase is a cytochrome P450 enzyme. In addition, enzymatic conversion of (+)-costunolide into 11(S),13-dihydrocostunolide and leucodin, a guaianolide, was detected. The first-mentioned reaction involves an enoate reductase, whereas the formation of leucodin from (+)-costunolide probably involves more than one enzyme, including a cytochrome P450 enzyme.     

Amorpha-4,11-diene synthase catalyses the first probable step in artemisinin biosynthesis

The endoperoxide sesquiterpene lactone artemisinin and its derivatives are a promising new group of drugs against malaria. Artemisinin is a constituent of the annual herb Artemisia annua L. So far only the later steps in artemisinin biosynthesis--from artemisinic acid--have been elucidated and the expected olefinic sesquiterpene intermediate has never been demonstrated. In pentane extracts of A. annua leaves we detected a sesquiterpene with the mass spectrum of amorpha-4,11-diene. Synthesis of amorpha-4,11-diene from artemisinic acid confirmed the identity. In addition we identified several sesquiterpene synthases of which one of the major activities catalysed the formation of amorpha-4,11-diene from farnesyl diphosphate. This enzyme was partially purified and shows the typical characteristics of sesquiterpene synthases, such as a broad pH optimum around 6.5-7.0, a molecular mass of 56 kDa, and a K(m) of 0.6 microM. The structure and configuration of amorpha-4,11-diene, its low content in A. annua and the high activity of amorpha-4,11-diene synthase all support that amorpha-4,11-diene is the likely olefinic sesquiterpene intermediate in the biosynthesis of artemisinin.     

Biosynthesis of germacrene A carboxylic acid in chicory roots. Demonstration of a cytochrome P450 (+)-germacrene a hydroxylase and NADP+-dependent sesquiterpenoid dehydrogenase(s) involved in sesquiterpene lactone biosynthesis

Sprouts of chicory (Cichorium intybus), a vegetable grown in the dark, have a slightly bitter taste associated with the presence of guaianolides, eudesmanolides, and germacranolides. The committed step in the biosynthesis of these compounds is catalyzed by a (+)-germacrene A synthase. Formation of the lactone ring is the postulated next step in biosynthesis of the germacrene-derived sesquiterpene lactones. The present study confirms this hypothesis by isolation of enzyme activities from chicory roots that introduce a carboxylic acid function in the germacrene A isopropenyl side chain, which is necessary for lactone ring formation. (+)-germacrene A is hydroxylated to germacra-1(10),4,11(13)-trien-12-ol by a cytochrome P450 enzyme, and is subsequently oxidized to germacra-1(10),4,11(13)-trien-12-oic acid by NADP+-dependent dehydrogenase(s). Both oxidized germacrenes were detected as their Cope-rearrangement products elema-1,3,11(13)-trien-12-ol and elema-1,3,11(13)-trien-12-oic acid, respectively. The cyclization products of germacra-1(10),4,11(13)-trien-12-ol, i.e. costol, were also observed. The (+)-germacrene A hydroxylase is inhibited by carbon monoxide (blue-light reversible), has an optimum pH at 8.0, and hydroxylates beta-elemene with a modest degree of enantioselectivity.     

Localization of enzymes of artemisinin biosynthesis to the apical cells of glandular secretory trichomes of Artemisia annua L.

A method based on the laser microdissection pressure catapulting technique has been developed for isolation of whole intact cells. Using a modified tissue preparation method, one outer pair of apical cells and two pairs of sub-apical, chloroplast-containing cells, were isolated from glandular secretory trichomes of Artemisia annua. A. annua is the source of the widely used antimalarial drug artemisinin. The biosynthesis of artemisinin has been proposed to be located to the glandular trichomes. The first committed steps in the conversion of FPP to artemisinin are conducted by amorpha-4,11-diene synthase, amorpha-4,11-diene hydroxylase, a cytochrome P450 monooxygenase (CYP71AV1) and artemisinic aldehyde Δ11(13) reductase. The expression of the three biosynthetic enzymes in the different cell types has been studied. In addition, the expression of farnesyldiphosphate synthase producing the precursor of artemisinin has been investigated. Our experiments showed expression of farnesyldiphosphate synthase in apical and sub-apical cells as well as in mesophyl cells while the three enzymes involved in artemisinin biosynthesis were expressed only in the apical cells. Elongation factor 1α was used as control and it was expressed in all cell types. We conclude that artemisinin biosynthesis is taking place in the two outer apical cells while the two pairs of chloroplast-containing cells have other functions in the overall metabolism of glandular trichomes.     

Valencene oxidase CYP706M1 from Alaska cedar (Callitropsis nootkatensis)

(+)-Nootkatone is a natural sesquiterpene ketone used in grapefruit and citrus flavour compositions. It occurs in small amounts in grapefruit and is a major component of Alaska cedar (Callitropsis nootkatensis) heartwood essential oil. Upon co-expression of candidate cytochrome P450 enzymes from Alaska cedar in yeast with a valencene synthase, a C. nootkatensis valencene oxidase (CnVO) was identified to produce trans-nootkatol and (+)-nootkatone. Formation of (+)-nootkatone was detected at 144 ± 10 μg/L yeast culture. CnVO belongs to a new subfamily of the CYP706 family of cytochrome P450 oxidases.     

Coupling cell growth and biochemical pathway induction in Saccharomyces cerevisiae for production of (+)-valencene and its chemical conversion to (+)-nootkatone

(+)-Nootkatone is a valuable, functional sesquiterpene that is widely used in food, cosmetics, pharmaceutical, agriculture, and other fields. However, only traces of it accumulate in plants, which is insufficient to meet the market demand. Therefore, commercial (+)-nootkatone is currently synthesized from (+)-valencene. Here, we engineered Saccharomyces cerevisiae to achieve high production of (+)-valencene. Employing gene screening, protein engineering and biosynthetic pathway optimization, we achieved 12.4 g/L (+)-valencene production with the mutant strain. This titer was further increased to 16.6 g/L, the highest titer reported to date, by coupling critical factors for cell growth and biochemical pathway induction. Subsequently, (+)-nootkatone was chemically synthesized from bio-fermented (+)-valencene with a yield of 80%. This study achieved efficient microbial synthesis of (+)-valencene, which may be utilized in industrial production and stabilize the supply of (+)-nootkatone.     

Salicylic acid activates artemisinin biosynthesis in <Emphasis Type="Italic">Artemisia annua</Emphasis> L.

This paper provides evidence that salicylic acid (SA) can activate artemisinin biosynthesis in Artemisia annua L. Exogenous application of SA to A. annua leaves was followed by a burst of reactive oxygen species (ROS) and the conversion of dihydroartemisinic acid into artemisinin. In the 24 h after application, SA application led to a gradual increase in the expression of the 3-hydroxy-3-methylglutaryl coenzyme A reductase (HMGR) gene and a temporary peak in the expression of the amorpha-4,11-diene synthase (ADS) gene. However, the expression of the farnesyl diphosphate synthase (FDS) gene and the cytochrome P450 monooxygenase (CYP71AV1) gene showed little change. At 96 h after SA (1.0 mM) treatment, the concentration of artemisinin, artemisinic acid and dihydroartemisinic acid were 54, 127 and 72% higher than that of the control, respectively. Taken together, these results suggest that SA induces artemisinin biosynthesis in at least two ways: by increasing the conversion of dihydroartemisinic acid into artemisinin caused by the burst of ROS, and by up-regulating the expression of genes involved in artemisinin biosynthesis.     

Bioconversion of (+)-valencene in submerged cultures of the ascomycete Chaetomium globosum

Submerged cultures of the ascomycete Chaetomium globosum oxidised the exogenous sesquiterpene (+)-valencene to nootkatone via the stereoselective generation of -nootkatol. Inhibition experiments suggested that the first introduction of oxygen, the rate-limiting step of the bioconversion, may have been catalysed by a cytochrome-P450-monooxygenase. However, nootkatone was not the final metabolite: further flavour-active and inactive, non-volatile oxidation products were identified. (+)-Valencene and the flavour-active mono-oxyfunctionalised transformation products, -nootkatol, nootkatone, and valencene-11,12-epoxide accumulated preferably inside the fungal cells. Di- and poly-oxygenated products, such as nootkatone-11,12-epoxide, were found solely in the culture medium, indicating an active transport of these metabolites into the extracellular compartment during (+)-valencene detoxification. These metabolic properties may have contributed to the high tolerance of the fungus towards the exogenous hydrocarbon.     

Molecular cloning, expression, and characterization of amorpha-4,11-diene synthase, a key enzyme of artemisinin biosynthesis in Artemisia annua L

In plants, sesquiterpenes of different structural types are biosynthesized from the isoprenoid intermediate farnesyl diphosphate. The initial reaction of the biosynthesis is catalyzed by sesquiterpene cyclases (synthases). In Artemisia annua L. (annual wormwood), a number of such sesquiterpene cyclases are active. We have isolated a cDNA clone encoding one of these, amorpha-4,11-diene synthase, a putative key enzyme of artemisinin biosynthesis. This clone contains a 1641-bp open reading frame coding for 546 amino acids (63.9 kDa), a 12-bp 5'-untranslated end, and a 427-bp 3'-untranslated sequence. The deduced amino acid sequence is 32 to 51% identical with the sequence of other known sesquiterpene cyclases from angiosperms. When expressed in Escherichia coli, the recombinant enzyme catalyzed the formation of both olefinic (97.5%) and oxygenated (2.5%) sesquiterpenes from farnesyl diphosphate. GC-MS analysis identified the olefins as (E)-beta-farnesene (0.8%), amorpha-4,11diene (91.2%), amorpha-4,7(11)-diene (3.7%), gamma-humulene (1.0%), beta-sesquiphellandrene (0.5%), and an unknown olefin (0.2%) and the oxygenated sesquiterpenes as amorpha-4-en-11-ol (0.2%) (tentatively), amorpha-4-en-7-ol (2.1%), and alpha-bisabolol (0.3%) (tentatively). Using geranyl diphosphate as substrate, amorpha-4,11-diene synthase did not produce any monoterpenes. The recombinant enzyme has a broad pH optimum between 7.5 and 9.0 and the Km values for farnesyl diphosphate, Mg2+, and Mn2+ are 0.9, 70, and 13 microM, respectively, at pH 7.5. A putative reaction mechanism for amorpha-4,11-diene synthase is suggested.     

在大肠杆菌内引入甲羟戊酸途径高效合成抗疟药青蒿素前体——紫穗槐-4,11-二烯

以青蒿素为基础的联合药物疗法 (ACTs) 被认为是目前治疗恶性疟疾的最有效方法。然而青蒿素供应不足且价格昂贵,限制了ACTs的广泛使用。采用基因工程手段构建异源类异戊二烯生物合成途径,利用大肠杆菌发酵能高效合成抗疟药青蒿素前体——紫穗槐-4,11-二烯。首先在大肠杆菌Escherichia coli DHGT7中引入人工合成的紫穗槐-4,11-二烯合酶基因,利用大肠杆菌内源的法尼基焦磷酸,成功获得了紫穗槐-4,11-二烯。为提高前体供给,引入粪肠球菌的甲羟戊酸途径,紫穗槐-4,11-二烯的产量提高了13     

Expression of a synthetic Artemesia annua amorphadiene synthase in Aspergillus nidulans yields altered product distribution

A gene encoding a plant terpene cyclase, Artemisia annua amorpha-4,11-diene synthase (ADS), was expressed in Aspergillus nidulans under control of a strong constitutive promoter, (p)gpdA. The transformants produced only small amounts of amorphadiene, but much larger amounts of similar sesquiterpenes normally produced as minor by-products in planta. In contrast, expression of ADS in Escherichia coli produced almost exclusively amorpha-4,11-diene. These results indicate that the host environment can greatly impact the terpenes produced from terpene synthases.     

Enhanced artemisinin production from engineered yeast precursors upon biotransformation

Abstract

Production of artemisinin in genetically modified microorganisms is an attractive option to enable sufficient supply of the effective antimalarial agent. Although a sundry of artemisinin precursors are available from engineered bacteria or yeast, no artemisinin has been manufactured by engineering any microbial platforms due to inaccessibility to unidentified steps. To this end, it is essential to consider how to convert artemisinin precursors to artemisinin, either biochemically or chemically. To establish a novel procedure of artemisinin production, we incubate the mixture of artemisinin precursors from engineered Sacchromyces cerevisiae with the cell-free enzyme extract of Artemisia annua. For the single gene-expressing strain INVScI (pYES-ADS), amorpha-4,11-diene accumulation within 48 h or 14 days led to higher artemisinin content than the control. In the multiple gene-expressing strain YPH501 (pYES-ADS:: pESC-CYP71AV1-DBR2), artemisinin accumulation from the 14-day-induced yeast precursor mixture was nearly equivalent between the single gene-transferred strain and the multiple gene-transferred strain. Alternatively, biotransformation of 48-hour-induced yeast amorpha-4,11-diene mixture by the cold-acclimated A. annua cell-free extract that possesses the abundant enzymes relevant to artemisinin biosynthesis gave rise to considerable elevation of artemisinin content up to 0.647% in maximum, accounting to 15-folds increase as the A. annua cell-free extract without cold-acclimation (0.045%), thereby providing a practical protocol for artemisinin overproduction through the interplay of engineered microbial artemisinin precursors with upregulated plant enzymes.     

Expression, purification, and characterization of recombinant amorpha-4,11-diene synthase from Artemisia annua L

A cDNA clone encoding amorpha-4,11-diene synthase from Artemisia annua was subcloned into a bacterial expression vector in frame with a His6-tag. Recombinant amorpha-4,11-diene synthase was produced in Escherichia coli and purified to apparent homogeneity. The enzyme showed pH optimum at pH 6.5, and a minimum at pH 7.5. Substantial activity was observed in the presence of Mg2+, Mn2+ or Co2+ as cofactor. The enzyme exhibits a low activity in the presence of Ni2+ and essentially no activity with Cu2+ or Zn2+. The sesquiterpenoids produced from farnesyl diphosphate in the presence of Mg2+ were analyzed by GC-MS. In addition to amorpha-4,11-diene, 15 sesquiterpenoids were produced. Only small quantitative differences in product pattern were observed at pH 6.5, 7.5, or 9.5. Amorpha-4,11-diene synthase showed significant increased product selectivity in the presence of Mn2+ or Co2+. Km for farnesyl diphosphate was 3.3, 8.0, and 0.7 microM in the presence of Mg2+, Mn2+ or Co2+, respectively. The corresponding kcat-values were 6.8, 15.0, and 1.3 x 10(-3) s(-1), respectively. Km and kcat for geranyl diphosphate were 16.9 microM and 7.0 x 10(-4) s(-1), respectively, at pH 6.5, in the presence of Mn2+.     

Nootkatone—a biotechnological challenge

Due to its pleasant grapefruit-like aroma and various further interesting molecular characteristics, (+)-nootkatone represents a highly sought-after specialty chemical. (+)-Nootkatone is accumulated in its producer plants in trace amounts only, and the demand of the food, cosmetics and pharmaceutical industry is currently predominantly met by chemical syntheses. These typically require environmentally critical reagents, catalysts and solvents, and the final product must not be marketed as a “natural flavour” compound. Both the market pull and the technological push have thus inspired biotechnologists to open up more attractive routes towards natural (+)-nootkatone. The multifaceted approaches for the de novo biosynthesis or the biotransformation of the precursor (+)-valencene to (+)-nootkatone are reviewed. Whole-cell systems of bacteria, filamentous fungi and plants, cell extracts or purified enzymes have been employed. A prominent biocatalytic route is the allylic oxidation of (+)-valencene. It allows the production of natural (+)-nootkatone in high yields under mild reaction conditions. The first sequence data of (+)-valencene-converting activities have just become known.

Challenges and pitfalls of P450-dependent (+)-valencene bioconversion by Saccharomyces cerevisiae

Natural nootkatone is a high value ingredient for the flavor and fragrance industry because of its grapefruit flavor/odor, low sensorial threshold and low availability. Valencene conversion into nootkatol and nootkatone is known to be catalyzed by cytochrome P450 enzymes from both prokaryotic and eukaryotic organisms, but so far development of a viable bioconversion process using either native microorganisms or recombinant enzymes was not successful. Using an in silico gene-mining approach, we selected 4 potential candidate P450 enzymes from higher plants and identified two of them that selectively converted (+)-valencene into β-nootkatol with high efficiency when tested using recombinant yeast microsomes in vitro. Recombinant yeast expressing CYP71D51v2 from tobacco and a P450 reductase from arabidopsis was used for optimization of a bioconversion process. Bioconversion assays led to production of β-nootkatol and nootkatone, but with low yields that decreased upon increase of the substrate concentration. The reasons for this low bioconversion efficiency were further investigated and several factors potentially hampering industry-compatible valencene bioconversion were identified. One is the toxicity of the products for yeast at concentrations exceeding 100 mg L⁻¹. The second is the accumulation of β-nootkatol in yeast endomembranes. The third is the inhibition of the CYP71D51v2 hydroxylation reaction by the products. Furthermore, we observed that the formation of nootkatone from β-nootkatol is not P450-dependent but catalyzed by a yeast component. Based on these data, we propose new strategies for implementation of a viable P450-based bioconversion process.     

Nicotiana benthamiana as a production platform for artemisinin precursors

Background

Production of pharmaceuticals in plants provides an alternative for chemical synthesis, fermentation or natural sources. Nicotiana benthamiana is deployed at commercial scale for production of therapeutic proteins. Here the potential of this plant is explored for rapid production of precursors of artemisinin, a sesquiterpenoid compound that is used for malaria treatment.

Methodology/Principal Findings

Biosynthetic genes leading to artemisinic acid, a precursor of artemisinin, were combined and expressed in N. benthamiana by agro-infiltration. The first committed precursor of artemisinin, amorpha-4,11-diene, was produced upon infiltration of a construct containing amorpha-4,11-diene synthase, accompanied by 3-hydroxy-3-methylglutaryl-CoA reductase and farnesyl diphosphate synthase. Amorpha-4,11-diene was detected both in extracts and in the headspace of the N. benthamiana leaves. When the amorphadiene oxidase CYP71AV1 was co-infiltrated with the amorphadiene-synthesizing construct, the amorpha-4,11-diene levels strongly decreased, suggesting it was oxidized. Surprisingly, no anticipated oxidation products, such as artemisinic acid, were detected upon GC-MS analysis. However, analysis of leaf extracts with a non-targeted metabolomics approach, using LC-QTOF-MS, revealed the presence of another compound, which was identified as artemisinic acid-12-β-diglucoside. This compound accumulated to 39.5 mg.kg⁻¹ fwt. Apparently the product of the heterologous pathway that was introduced, artemisinic acid, is further metabolized efficiently by glycosyl transferases that are endogenous to N. benthamiana.

Conclusion/Significance

This work shows that agroinfiltration of N. bentamiana can be used as a model to study the production of sesquiterpenoid pharmaceutical compounds. The interaction between the ectopically introduced pathway and the endogenous metabolism of the plant is discussed.