Introduction!

Artemisinin, a sesquiterpene lactone endoperox-ide, is an effective antimalarial drug extractedfrom the leaves of the Chinese medicinal herb Ar-temisia annua L. (Asteraceae) [1,2]. Because of theincreasing resistance of Plasmodium falciparum tothe traditional antimalarial drugs (quinine andchloroquine), artemisinin and its derivatives havebecome the most important agents in the treat-ment of malaria, particularly in the form of arte-misinin-based combination therapies (ACTs). Atpresent, the only commercial source of artemisi-nin comes from the extraction of naturalA. annua, but the low content (0.01–0.8%) of arte-misinin in A. annua makes it a relatively expen-

sive drug, especially for the malaria-prone Africancountries. In order to make artemisinin-basedcombination therapies affordable, considerableefforts have been taken to increase the artemisi-nin content of A. annua or to develop more effi-cient ways of artemisinin production.The biosynthetic pathway of artemisinin belongsto the isoprenoid pathway. In this pathway(l" Fig. 1), farnesyl diphosphate (FPP) occupies acentral position, from which isoprene units aredispatched to the different classes of isoprenoidswith versatile structures and functions. InA. annua, FPP serves as a common precursor of ar-temisinin diverted by amorpha-4,11-diene syn-thase (ADS) and of other sesquiterpenes divertedby different sesquiterpene synthases through

Abstract!

Artemisinin is an effective antimalarial drug iso-lated from the medicinal plant Artemisia annua L.Due to its increasing market demand and the lowyield in A. annua, there is a great interest in in-creasing its production. In this paper, in an at-tempt to increase artemisinin content ofA. annua by suppressing the expression of β-car-yophyllene synthase, a sesquiterpene synthasecompeting as a precursor of artemisinin, the anti-sense fragment (750 bp) of β-caryophyllene syn-thase cDNA was inserted into the plant expres-sion vector pBI121 and introduced into A. annuaby Agrobacterium-mediated transformation. PCRand Southern hybridization confirmed the stableintegration of multiple copies of the transgene in5 different transgenic lines of A. annua. Reversetranscription PCR showed that the expression ofendogenous CPS in the transgenic lineswas signif-icantly lower than that in the wild-type controlA. annua plants, and β-caryophyllene content de-creased sharply in the transgenic lines in compar-ison to the control. The artemisinin content of one

of the transgenic lines showed an increase of54.9% compared with the wild-type control. Thepresent study demonstrated that the inhibitionpathway in the precursor competition for artemi-sinin biosynthesis by anti-sense technology is aneffectivemeans of increasing the artemisinin con-tent of A. annua plants.

Abbreviations!

ALDH1: aldehyde dehydrogenase 16-BA: 6-benzyl aminopurineCPS: β-caryophyllene synthaseCefo: cefotaxineCYP71AV1: cytochrome P450 monooxygenaseDBR2: double bond reductase 2FPS: fanesyl diphosphate synthaseHMGR: 3-hydroxy-3-methylglutary-CoA

reductaseKan: kanamycininNAA: α-naphthalene acetic acidRED1: reductase 1

Artemisinin Biosynthesis Enhancement in TransgenicArtemisia annua Plants by Downregulation of theβ-Caryophyllene Synthase Gene

Authors Jian-Lin Chen1,2, Hua-Ming Fang2, Yun-Peng Ji1,2, Gao-Bin Pu3, Yan-Wu Guo2, Li-Li Huang1,2, Zhi-Gao Du2,Ben-Ye Liu2, He-Chun Ye2, Guo-Feng Li2, Hong Wang1,2

Affiliations 1 Graduate University of the Chinese Academy of Sciences, Beijing, China2 Key Laboratory of Photosynthesis and Environmental Molecular Physiology, Institute of Botany,The Chinese Academy of Sciences, Beijing, China

3 Shandong Yingcai University, Jinan, China

Key wordsl" Artemisia annua L.l" Asteraceael" artemisinin biosynthesisl" anti‑sense β‑caryophyllene

synthase genel" genetic transformation

received August 9, 2010revised March 11, 2011accepted March 30, 2011

BibliographyDOI http://dx.doi.org/10.1055/s-0030-1271038Published online April 20, 2011Planta Med 2011; 77:1759–1765 © Georg ThiemeVerlag KG Stuttgart · New York ·ISSN 0032‑0943

CorrespondenceDr. Hong WangCollege of Life SciencesGraduate Universityof the Chinese Academyof Sciences19A YuquanluBeijing, 100049P.R. ChinaPhone: + 861062836239Fax: + 861088256080wangh@ibcas.ac.cn

1759

Chen J-L et al. Artemisinin Biosynthesis Enhancement… Planta Med 2011; 77: 1759–1765

Original Papers

competitive pathways. Therefore, the conversion step catalyzedby sesquiterpene synthases represents a metabolic shuntingpoint at which the cellular carbon flux is directed towards arte-misinin or other sesquiterpenes, depending on their competitionwith the available FPP pool. A number of cDNAs that encode dif-ferent sesquiterpene synthases have been isolated in recentyears. In 1999, the first cDNA clone encoding a sesquiterpenesynthase in A. annua, epi-cedrol synthase, was isolated [3,4]. Lat-er, 4 cDNAs encoding amorpha-4,11-diene synthase [5–7], β-car-yophyllene synthase [8], (E)-β-farnesene synthase [9], and ger-macrene A synthase [10] were isolated from A. annua. Amongthe five above-mentioned sesquiterpene synthases, except ADS,all the other four sesquiterpene synthases compete with FPP inartemisinin biosynthesis. So, if the gene expression of the sesqui-terpene synthase(s) competing with FPP in artemisinin biosyn-thesis could be blocked by antisense technology or RNA interfer-ence, then the FPP contribution to artemisinin biosynthesis couldhelpfully be increased, and the biosynthesis of artemisinin couldbe enhanced.In the present study, the antisense fragment of β-caryophyllenesynthase cDNA (asCPS) was introduced into A. annua by an Agro-bacterium-mediated transformation approach to assess the ef-fects of inhibiting the competitive pathway on artemisinin bio-synthesis. In addition, the changes in artemisinin, caryophyllene,and other related metabolites were investigated.

Materials and Methods!

Plant material and culture conditionsA. annua L. seeds of the high artemisinin yielding strain 001 werecollected from Sichuan Province, China. The seeds used in thisstudy were identified by Prof. Yi-Lin Chen (Institute of Botany,the Chinese Academy of Sciences). They were surface-sterilized

with 0.1% mercuric chloride for 10min, followed by washingthoroughly with sterile distilled water. Then the seeds were ger-minated on MS medium supplemented with 3.0% sucrose andsolidified with 0.7% agar. The cultures were incubated at 26°Cunder 16-h photoperiods. Germination started within 2 or 3days. Four weeks later, the seedlings were transferred to MS me-dium, and the leaves of 3-week-old seedlings were used as ex-plants for genetic transformation.

Construction of plasmids and transformationof Agrobacterium tumefaciensThe plasmid containing A. annua β-caryophyllene synthase cDNA[8], kindly provided by Prof. Xiao-Ya Chen (Institute of PlantPhysiology and Ecology, Shanghai Institutes for Biological Sci-ences, the Chinese Academy of Sciences), was used as the tem-plate for the amplification of the CPS fragment, then the ampli-fied fragment was used for the construction of an antisense plantexpression vector. The PCR primers were designed according tothe sequence at the 5′ and 3′ terminal region of CPS: 5′-CTGAGCTCTCCCAAACAATCTACCTTATG‑3′ (SacI restriction siteunderlined) and 5′-CTGGATCCTCGGAATATCTTTACGCACAA‑3′(BamHI restriction site underlined). The PCR was performed inan Eppendorf Mastercycler thermocycler under the followingconditions: 5min at 94°C; 30 cycles of reactions by 40 s at 94°C,40 s at 57°C, 1min 40 s at 72°C; plus a final 10min at 72°C. Theamplified CPS fragment of 750 bp length was collected after di-gestion with SacI and BamHI (Takara). The binary vector pBI121,containing both the NPT II gene controlled by the NOS promoterand the GUS gene controlled by the cauliflower mosaic virus(CaMV) 35S promoter, was also digestedwith SacI and BamHI (Ta-kara), and the resulting large fragment was collected. The abovetwo collected fragments were fused with T4 DNA ligase (Takara)to generate the recombinant plasmid pBIasCPS (l" Fig. 2).

Fig. 1 Diagram of the mevalonic pathway leadingto the biosynthesis of artemisinin and other isopre-noids. ADS: amorpha-4,11-diene synthase; CPS:β-caryophyllene synthase; CYP71AV1: cytochromeP450 monooxygenase; DBR2: double bond reduc-tase 2; HMG‑CoA: 3-hydroxy-3-methylglutary-CoA;HMGR: 3-hydroxy-3-methylglutary-CoA reductase;ECS: epi-cedrol synthase; FPP: fanesyl diphosphate;FPS: fanesyl diphosphate synthase; FS: (E)-β-farne-sene synthase; GS: germacrene A synthase; IPP: iso-pentenyl diphosphate; MVA: mevalonic acid; RED1:reductase 1; SQS: squalene synthase.

1760

Chen J-L et al. Artemisinin Biosynthesis Enhancement… Planta Med 2011; 77: 1759–1765

Original Papers

The recombinant plasmid pBIasCPS was amplified in DH5α andthen introduced into A. tumefaciens strain EHA105 by thefreeze-thawmethod. Transformants were selected on LBmedium(pepton 10 g, yeast extract 1 g, NaCl 5 g, pH 7.0) supplementedwith 50mg/L Kan. A. tumefaciens harboring the recombinant vec-tor pBIasCPS was used for A. annua transformation.

Transformation of A. annuaThe transformation of A. annua was performed essentially ac-cording to the method of Han et al. [11]. Young leaves of the seed-lings subcultured for about 3 weeks were used for the transfor-mation experiments. Prior to infection, a monoclone ofA. tumefaciens strain EHA105 harboring the recombinant vectorpBIasCPS was picked out and incubated at 28°C under agitationfor 36 h, then 1mL bacterial suspension was diluted to 50mL LBliquid medium and incubated under the same conditions tillOD600 reached 0.6–0.8. Afterwards, the bacterial suspension wascentrifuged, resuspended with liquid MS medium, followed byincubation of another 3 h for further activation, then the acti-vated bacterial suspension was used for A. annua transformation.The wounded leaves were immersed in the bacterial suspensionfor 20min, blotted on sterile filter paper and then cocultivated onMS medium in dark at 28°C for 2 or 3 days. After cocultivation,the leaves were transferred to shoot inducing medium (MS me-dium supplemented with 1.0mg/L 6-BA, 0.05mg/L NAA,500mg/L Cefo, and 15mg/L Kan) to induce shoots. Cefo was usedto eliminate surplus Agrobacterium; Kan was used to select thetransformants, and the media without Kan were considered ascontrols.Seven days later, the leaves were transferred to fresh shoot induc-ing medium with 400mg/L Cefo. Afterwards, the medium waschanged every 10 days with gradually decreased concentrationsof Cefo. One month later, the shoots were transferred to 100mLErlenmeyer flasks containing 40mL MS medium with 250mg/LCefo and subcultured every 3 weeks. When the induced shootshad grown about 2 cm height, the shoots were separated fromthe explants and transferred to root inducing medium (MS me-dium supplemented with 0.05mg/L NAA and 15mg/L Kan) to in-duce roots. Our voucher specimens were deposited at the In VitroPlant Germplasm Collection (IVPGC) of the Chinese Academy ofSciences, under the registration number A-annua-001 (control)and A-annua-CPS (asCPS transgenic plant).

Isolation of genomic DNA and PCR analysisThe genomic DNA of Kan resistant A. annua and control was iso-lated using the modified CTAB method [12]. The forward primerwas 5′-CCACGTCTTCAAAGCAAGTGGATTG‑3′, designed accordingto the sequence of the CaMV35S promoter. The reverse primerwas 5′-CTGAGCTCTCCCAAACAATCTACCTTATG-3′, designed ac-cording to the sequence of CPS. The 20 µL PCR system contained10 pmol forward primer, 10 pmol reverse primer, 50 ng plant ge-nomic DNA as template, 2 µL 10 × Taq DNA polymerase buffer,

0.25mmol/L dNTPs, and 1U Taq DNA polymerase (Takara). Am-plification was performed in a thermal cycler (Mastercycler; Ep-pendorf) as: 5min at 94°C; 30 cycles of reactions by 50 s at 94°C,1min at 56°C, and 1min at 72°C; plus a final 10min at 72°C. Inthe PCR analyses, DNA from plasmid pBIasCPS and that from thenon-transformed plants were used as the positive control andthe negative control, respectively. The amplification fragmentswere electrophoresed on a 1% agarose gel containing ethidiumbromide and photographed.

Southern blot analysisGenomic DNA (10 µg) was digested with HindIII (Takara) over-night, then electrophoresed on a 0.7% agarose gel. After electro-phoresing, the DNA was transferred onto a Hybond-N+ nylonmembrane (Amersham Bioscience) by the capillary transfermethod.The probe was obtained by means of PCR and then purified usingthe TIAN gel Midi Purification Kit (Tiangen Biotech). The primersand PCR conditions were the same as those in the above PCRanalyses. The probe was purified and then labeled by [α-32P]dCTP. The probe preparation was carried out according to theprotocol of the Prime-a-Gene Labeling System (Promega). Thehybridization and washing membrane were performed accord-ing to Sambrook et al. [13].

Isolation of total RNA and RT‑PCR analysisLeaves of the transgenic plants and control were collected for to-tal RNA isolation. Total RNAwas isolated by a commercially avail-able Plant RNA Kit (Autolabtech). The reverse transcription wasperformed with the random primers and by the TIANScript RTKit (Tiangen Biotech). Briefly, 1 µg of total RNA, 0.2 µg of randomprimers, and 2 µL dNTP were mixed, and DEPC water was supple-mented to 13.5 µL of total volume. The mixture was incubated at70°C for 5min, then rapidly placed in an ice bath for 2min. Then,4 µL of 5 × First-Strand buffer, 0.5 µL RNasin, and 1 µL DTT wereadded to each sample and incubated at 25°C for 10min. Afteradding 1 µL of M–MLV (200U), the reaction mixture with 20 µLof the total volume was centrifuged and incubated first at 42°Cfor 50min, then at 95°C for 5min. The prepared cDNAwas storedat −20°C until use.The primers and PCR conditions used in the RT‑PCR analyses arelisted in l" Table 1.

Detection of artemisinin using HPLCThe detection of artemisinin was performed according to themethod of Zhao et al. [14]. The leaves of transgenic and controlplants were collected and dried to constant weight in a 50°Coven. Then the dried leaves were ground into fine powder. Ap-proximately 40mL of petroleum ether (boiling range 30–60°C)were mixed with 50mg of the sample powder for artemisinin ex-traction by the sonication method. After filtrating and drying, theextract was transferred into a flask and diluted by methanol to

Fig. 2 Diagram of the plant expression vectorpBIasCPS.

1761

Chen J-L et al. Artemisinin Biosynthesis Enhancement… Planta Med 2011; 77: 1759–1765

Original Papers

10mL of the total volume. Then 1mL of the solution was mixedwith 4mL 0.2% (W/V) NaOH and then incubated at 50°C for 30min. After being cooled to room temperature, 5mL 0.05mol/Lacetic acid was added to the mixture, then filtrated with a0.45 µm disposable syringe filter.10 µL of the sample solutionwas injected into the sampling outletof the HPLC. The tests were carried out in a Kromosil C18 reversecolumn (4.6mm×250mm, 5 µmparticle size). The 0.01mol/L so-dium phosphate buffer (pH 7.0) –methanol mixture (55:45, V/V)was used as a mobile phase with a 1mL/min flow rate, 30°C col-umn temperature, and 260 nmwavelength for artemisinin detec-tion.A reference standard of artemisinin was dissolved in methanoland prepared in the same way as the sample. The standard curvewas prepared by the external reference method. The regressionequation and correlation coefficient (R2) of each standard curvewere automatically determined with an EXCEL software system.The R2 of the regression equation for artemisinin was 0.9835, in-dicating a high linearity correlation for the artemisinin standardcurve.

Determination of secondary metabolites relatedto artemisinin biosynthesisLeaves of the transgenic plants and control were collected. Theleaves were dried in a 50°C oven for artemisinin detection andin a freeze dryer for caryophyllene, artemisinic acid, and dihy-droartemisinic acid detection; then the dried leaves were groundinto fine powder. Some 50mg of the exactly weighted powderwas mixed with 40mL hexane for the extraction by the sonica-tion method. After filtration, the extract was condensed to 1mLof the total volume.

One µL of the sample solution was used for GC (Varian 3800)analysis. Samples were quantitatively analyzed by the externalstandardmethod. The regression equation and correlation coeffi-cient (R2) of each standard curve were determined, with R2 rang-ing from 0.9627 to 0.9855, indicating that a high linearity corre-lation was achieved for all the standard curves.A capillary column DB-5 (30m× 0.25mm, film thickness of0.25 µm)was usedwith the inlet temperature at 250°C, detection(FID) at 290°C, split ratio of 1 :1, and flow rate at 1mL/min withhelium as the carrier gas.For caryophyllene detection (according to Rasmann et al. [15]),GC temperature programming started from 80°C, at 5°C/min to240°C, then at 20°C/min to 280°C, and stayed at 280°C for 5min. For the detection of AA and DHAA (according to Wallaart etal. [16]), GC temperature programming started from 180°C,stayed at 180°C for 10min, then increased to 280°C at 15°C/minand stayed at 280°C for 5min.Standard artemisinin (purity > 95%) and artemisinic acid (purity >95%) were kindly provided by Prof. Ying Li (Institute of Pharmacy,the Chinese Academy of Sciences, Shanghai, China). Dihydroarte-misinic acid was donated by Dr. Harro J. Bouwmeester (Plant Re-search International, Wageningen, Netherlands). Authentic β-car-yophyllene (purity > 98.5%) was purchased from Sigma.

Results and Discussion!

The young leaves of about 3-week-old A. annua seedlings wereused as explants; Agrobaterium tumefaciens strain EHA105, har-boring the recombinant plasmid pBIasCPS, was used for transfor-mation. After cocultivation, shoot-inducing, and root-inducingculture, Kan resistant plants were regenerated. Compared withthe untransformed control plants, the phenotype and growth ofthe asCPS transformed plants showed no significant changes (da-ta not shown).PCR analysis was conducted on the lines that were putativelytransformed with the asCPS gene by their resistance to Kan. Spe-cific products of the about 900 bp fragment were amplified in atleast five lines of the Kan resistant plants; the size of the bandwas the same as that when pBIasCPS was used as the positivecontrol, and no DNA band was detected in the untransformedcontrol plants (l" Fig. 3). To rule out the possible false positivebrought out by the remnants of Agrobacterium tumefaciens, aconfirmation PCR analysis was performedwith the non-insertion

Table 1 Primers and PCR conditions used for RT‑PCR analyses.

Gene Primers PCR conditions

CPS 5′-GGAACCATGGGTAGTTTGAAAGCAGTATTG‑3′5′-GCCTGGATCCCTTGACTCTCTCTTAACTAT‑3′

94°C for 40 s, 58°Cfor 40 s, 72°C for1min 30 s, 28 cycles

FPS 5′-ATGGGTAGCATCGATCTGAAAT‑3′5′-TTTGCCTCTTGTAGATTTTACC‑3′

94°C for 40 s, 54°Cfor 40 s, 72°C for1min 20 s, 28 cycles

ADS 5′-TCATATACTCATAGGATAAAC‑3′5′-GGAATGGCACTTACAGAAGA‑3′

94°C for 40 s, 54°Cfor 40 s, 72°C for1min 30 s, 30 cycles

ALDH1 5′-CACCATGAGCTCAGGAGCTAAT‑3′5′-TTAAAGCCACGGGGAATCATAT‑3′

94°C for 40 s, 56°Cfor 40 s, 72°C for1min 40 s, 28 cycles

DBR2 5′-CACCATGTCTGAAAAACCAACCTTG‑3′5′-GCTCATAAGATGCACCTTAATAAG‑3′

94°C for 40 s, 57°Cfor 40 s, 72°C for1min 30 s, 30 cycles

CYP71-AV1

5′- CTAGGATCCATGGCACTCTCACTGACCACT-3′5′-TACTCGAGTAAATACTAGAAACTTGGAACG‑3′

94°C for 40 s, 55°Cfor 40 s, 72°C for1min 40 s, 28 cycles

DBR2 5′-CACCATGTCTGAAAAACCAACCTTG‑3′5′-GCTCATAAGATGCACCTTAATAAG‑3′

94°C for 40 s, 57°Cfor 40 s, 72°C for1min 30 s, 30 cycles

ACTIN 5′-AACTGGGATGACATGGAGAAGATAT‑3′5′-TCACACTTCATGATGGAGTTGTAGG‑3′

94°C for 40 s, 56°Cfor 40 s, 72°C for1min, 28 cycles

RED1 5′-CGAAATATGTCGTCAGTTGGCCTC‑3′5′-GACCATCATCGGGCAACAAAGC‑3′

94°C for 40 s, 53°Cfor 40 s, 72°C for1min, 27 cycles

Fig. 3 PCR analysis of putative asCPS transgenic and control plants. NC:negative control (DNA from wild-type control plant as the template);T2‑T10: DNA from putative asCPS transgenic plants as the template;PC: positive control (DNA from plasmid pBIasCPS as the template);M: DNA molecular weight maker.

1762

Chen J-L et al. Artemisinin Biosynthesis Enhancement… Planta Med 2011; 77: 1759–1765

Original Papers

section of pBIasCPS as the primer and the genome DNA of thetransgenic plants as the template. No band was amplified in allthe transgenic plants which excluded the presence of any re-maining Agrobacterium tumefaciens in the transgenic plants (da-ta not shown).Five PCR positive transformed A. annua plants were further de-tected by Southern blot analysis. As shown in l" Fig. 4, both theasCPS transformed and the untransformed control plants con-tained a common band of the same size which represented theendogenous β-caryophyllene synthase gene. Compared with theuntransformed control plant, two or three additional bands ap-peared in the transgenic plants, and the size of such bands wasdifferent in different transgenic lines, which indicated that thesetransgenic plants were different lines. The results of Southernblot confirmed the successful integration of the asCPS gene intodifferent loci of the A. annua genome.In order to investigate the effects of the introduction of asCPS onthe expression of other genes in the transgenic A. annua, the ex-pression of several genes related to the artemisinin biosynthesis(CPS, HMGR, FPS, ADS, CYP71AV1, DBR2, ALDH1, and RED1) wasanalyzed by RT‑PCR. The conserved ACTIN gene was used as anendogenous control for gene expression. The RT‑PCR results

(l" Fig. 5) showed that the transcriptional expression of endoge-nous CPS in each of the transgenic lines was largely inhibited.Meanwhile, the expression of artemisinin-biosynthesis-relatedgenes (HMGR, FPS, ADS, CYP71AV1, DBR2, and ALDH1) in thetransgenic plants was enhanced at various degrees comparedwith that of the control. The underlying reason could be that thedepression of CPS activity in the transgenic plants led more FPPto divert to the branch of artemisinin biosynthesis, which in-duced high expression of the genes involved in this biosynthesis.From l" Fig. 5, it could be also observed that RED1, a dihydroarte-misinic aldehyde reductase gene which may have a negative im-pact on artemisinin production [17], was apparently restrainedin all transgenic lines. The downregulation of RED1 of the trans-genic plants could have led more dihydroartemisinic aldehyde todivert to artemisinin biosynthesis (l" Fig. 1), and as a result, thecontent of artemisinin of the transgenic plants was increased.To investigate whether the introduction of asCPS affects the bio-synthesis of β-caryophyllene and artemisinin, leaves of the asCPStransformed and untransformed control plants were collected todetect β-caryophyllene content and artemisinin content. Arte-misinin was detected by HPLC and β-caryophyllene by GC.The results of β-caryophyllene determination are shown inl" Fig. 6. It can be seen that all the transformed plants exhibitedsignificantly lower β-caryophyllene content than the untrans-formed control. Of the transgenic lines, T3 and T2 had a β-caryo-phyllene content reduction of 62.7% and 54.1%, respectively,compared to the control, whereas lines T9, T10, and T4 had a re-duction of 40–50% in comparison with the control. The above re-sults indicated that our antisense suppression was of high effi-ciency.As shown in l" Fig. 7, the artemisinin content of the transformedplants increased significantly compared to that of the untrans-formed control, with an increase of 54.9% in some transgeniclines. This showed that the expression inhibition of CPS could ef-fectively prevent it from competing for precursor of artemisinin,and consequently promote the biosynthesis of artemisinin intransgenic A. annua. Comparing l" Fig. 6 with l" Fig. 7, it can beseen that the decrease in caryophyllene content (l" Fig. 6) corre-lated negatively with the increase in artemisinin (l" Fig. 7).Artemisinic acid and dihydroartemisinic acid are supposed to beprecursors of artemisinin biosynthesis, so the content of artemis-

Fig. 4 Southern blot analysis of asCPS transgenic and control plants. CK:control (DNA from wild-type control plant); T2-T10: DNA from asCPS trans-genic plants.

Fig. 5 RT‑PCR analysis of the artemisinin-biosynthesis-related genes ofasCPS transgenic and control plants. CK: control (wild-type control plant);T2-T10: asCPS transgenic plants.

Fig. 6 The content of caryophyllene, AA, and DHAA in asCPS transgenicand control plants. The GC analysis was performed with three independentsamples and replicated three times, each value represented the mean ofthe three replicates. Error bars represent standard deviation;* 0.01≤ p < 0.05; ** p < 0.01. AA: artemisinic acid; DHAA: dihydro-artemisinic acid; CK: control; T2-T10: asCPS transgenic plants.

1763

Chen J-L et al. Artemisinin Biosynthesis Enhancement… Planta Med 2011; 77: 1759–1765

Original Papers

inic acid and dihydroartemisinic acid of the asCPS transformedand untransformed control plants were detected by GC. Asshown in l" Fig. 6, the content of artemisinic acid in all asCPStransgenic lines decreased at various degrees comparedwith thatof the untransformed control, but the content of dihydroartemis-inic acid increased significantly in all transgenic plants. The pos-sible reasons of the observed dihydroartemisinic acid increase inthe transgenic plants could be: on the one hand, since the expres-sion of the artemisinin-biosynthesis-related genes (HMGR, FPS,ADS, CYP71AV1, DBR2, and ALDH1) in the transgenic plants wasenhanced, more precursors for artemisinin biosynthesis couldbe provided. On the other hand, the significantly decreased ex-pression of RED1 in all transgenic lines could reduce the con-sumption of its substrate, dihydroartemisinic aldehyde, so in-creasing the precursors for artemisinin biosynthesis.Currently, the biosynthetic pathway of artemisinin is not com-pletely understood; most data support that dihydroartemisinicacid is the precursor of artemisinin biosynthesis, and then dihy-droartemisinic acid converts to artemisinin by a nonenzymaticreaction [18]. In our present study, the dihydroartemisinic acidcontent of lines T9 and T10 was about two times the one of thecontrol (l" Fig. 6), and the artemisinin content of these two lines(T9, T10) was also higher than the other transgenic lines(l" Fig. 7). Our results also indicated that dihydroartemisinic acidis the precursor of artemisinin biosynthesis.In recent years, much progress has been made on the artemisininbiosynthetic pathway investigation, and many genes have beenisolated from A. annua, including the genes of the key enzymesinvolved in the artemisinin biosynthesis such as FPS [19], ADS[20], CYP71AV1 [21], DBR2 [22], and ALDH1 [23], as well as genesof other enzymes relevant to the artemisinin biosynthesis, suchas the squalene synthase gene [24], β-caryophyllene synthasegene [8], and dihydroartemisinic aldehyde reductase (RED1)gene [17], whichmakes the artemisinin production increase pos-sible by metabolic engineering. The two basic strategies are ei-ther overexpressing the key enzymes involved in the biosynthesisof artemisinin or inhibiting the enzymes of the pathways com-peting for precursors of artemisinin. In this paper, the secondstrategy was adopted.In order to get high-artemisinin-yield transgenic A. annua, theantisense β-caryophyllene synthase fragment was transferred in-to A. annua by A. tumefaciens-mediated transformation. Our re-sults show that the introduction of asCPS can effectively preventit from competing for the precursors of artemisinin biosynthesis,and consequently promote this biosynthesis. Meanwhile, the

phenotype and growth of the transgenic plants did not exhibitany marked changes compared to the control. Our result is con-sistent with that in the literature [25,26] reporting that the arte-misinin content was increased by inhibiting the sterol pathway(another pathway competing precursor with artemisinin biosyn-thesis, l" Fig. 1) by RNA interference or antisense strategies. Insum, our results show that the inhibition pathway in the precur-sor competition for artemisinin biosynthesis is an effective andsuitable means of increasing artemisinin content of A. annuaplants.

Acknowledgements!

The authors would like to thank Prof. Xiao-Ya Chen, Institute ofPlant Physiology and Ecology, Shanghai Institutes for BiologicalSciences, the Chinese Academy of Sciences for the gift of β-caryo-phyllene synthase cDNA. We also give our thanks to Prof. Zhao-Lin Lv, Research Center of Beijing Forestry University, for his helpin the GC analyses. This work was partially funded by the Nation-al Natural Science Foundation of China (60773164) as well as by“the President Fund of GUCAS”.

References1 Cooperative Research Group on Artemisinin Structure. Artemisinin – anew sesquiterpene lactone. Chin Sci Bull 1977; 22: 142

2 Liu JM, Ni MY, Fan JF, Tu YY, Wu ZH, Wu YL, Zhou WS. Structure and re-action of artemisinin. Acta Chim Sin 1979; 37: 129–143

3 Hua L, Matsuda SP. The molecular cloning of 8-epicedrol synthase fromArtemisia annua L. Arch Biochem Biophys 1999; 369: 208–212

4 Mercke P, Crock J, Croteau R, Brodelius PE. Cloning, expression, and char-acterization of epi-cedrol synthase, a sesquiterpene cyclase from Arte-misia annua L. Arch Biochem Biophys 1999; 369: 213–222

5 Mercke P, Bengtsson M, Bouwmeester HJ, Posthumus MA, Brodelius PE.Molecular cloning, expression, and characterization of amorpha-4,11-diene synthase, a key enzyme of artemisinin biosynthesis in Artemisiaannua L. Arch Biochem Biophys 2000; 381: 173–180

6 Chang YJ, Song SH, Park SH, Kim SU. Amorpha-4,11-diene synthase ofArtemisia annua: cDNA isolation and bacterial expression of a terpenesynthase involved in artemisinin biosynthesis. Arch Biochem Biophys2000; 383: 178–184

7 Wallaart TE, Bouwmeester HJ, Hille J, Poppinga L, Maijers NC. Amorpha-4,11-diene synthase: cloning and functional expression of a key en-zyme in the biosynthetic pathway of the novel antimalarial drug arte-misinin. Planta 2001; 21: 460–465

8 Cai Y, Jia JW, Crock J, Lin ZX, Chen XY, Croteau R. A cDNA clone for beta-caryophyllene synthase from Artemisia annua. Phytochemistry 2002;61: 523–529

9 Picaud S, Brodelius M, Brodelius PE. Expression, purification and charac-terization of recombinant (E)-beta-farnesene synthase from Artemisiaannua. Phytochemistry 2005; 66: 961–967

10 Bouwmeester HJ, Kodde J, Verstappen FW, Altug IG, de Kraker JW, Wal-laart TE. Isolation and characterization of two germacrene A synthasecDNA clones from chicory. Plant Physiol 2002; 129: 134–144

11 Han JL, Wang H, Ye HC, Liu Y, Li ZQ, Zhang Y, Zhang YS, Yan F, Li GF. Highefficiency of genetic transformation and regeneration of Artemisia an-nua L. via Agrobacterium tumefaciens-mediated procedure. Plant Sci2005; 168: 73–80

12 Allen G, Flores-Vergara M, Krasynanski S, Kumar S, Thompson W. Amodified protocol for rapid DNA isolation from plant tissues using ce-tyltrimethylammonium bromide. Nat Protoc 2006; 1: 2320–2325

13 Sambrook J, Fritsch EF, Mariatis T.Molecular cloning: A laboratoryman-ual, 2nd edition. New York: Cold Spring Harbor Laboratory Press;1989: 483–490

14 Zhao SS, Zeng MY. Determination of Qinghaosu in Artemisia annua L. byhigh performance liquid chromatography. Chin J Pharm Anal 1986; 6:3–5

15 Rasmann S, Köllner TG, Degenhardt J, Hiltpold I, Toepfer S, Kuhlmann U,Gershenzon J, Turlings TC. Recruitment of entomopathogenic nema-todes by insect-damaged maize roots. Nature 2005; 434: 732–737

Fig. 7 Artemisinin content of asCPS transgenic and control plants. TheHPLC analysis was performed with three independent samples and repli-cated three times, each value represented the mean of the three replicates.Error bars represent standard deviation; * 0.01≤ p < 0.05; ** p < 0.01.CK: control; T2-T10: asCPS transgenic plants.

1764

Chen J-L et al. Artemisinin Biosynthesis Enhancement… Planta Med 2011; 77: 1759–1765

Original Papers

16 Wallaart TE, Uden WV, Lubberink HG, Woerdenbag HJ, Pras N, Quax WJ.Isolation and identification of dihydroartemisinic acid from Artemisiaannua and its possible role in the biosynthesis of artemisinin. J NatProd 1999; 62: 430–433

17 Rydén A-M, Ruyter-Spira C, Quax WJ, Osada H, Muranaka T, Kayser O,Bouwmeester H. The molecular cloning of dihydroartemisinic aldehydereductase and its implication in artemisinin biosynthesis in Artemisiaannua. Planta Med 2010; 76: 1178–1183

18 Covello PS, Teoh KH, Polichuk DR, Reed DW, Nowak G. Functional ge-nomics and the biosynthesis of artemisinin. Phytochemistry 2007; 68:1864–1871

19 Matsushita Y, Kang W, Charlwood BV. Cloning and analysis of a cDNAencoding farnesyl diphosphate synthase from Artemisia annua. Gene1996; 172: 207–209

20 Picaud S, Olofsson L, Brodelius M, Brodelius PE. Expression, purification,and characterization of recombinant amorpha-4,11-diene synthasefrom Artemisia annua L. Arch Biochem Biophys 2005; 436: 215–226

21 Teoh KH, Polichuk DR, Reed DW, Nowak G, Covello PS. Artemisia annua L.(Asteraceae) trichome-specific cDNAs reveal CYP71AV1, a cytochromeP450 with a key role in the biosynthesis of the antimalarial sesquiter-pene lactone artemisinin. FEBS Lett 2006; 580: 1411–1416

22 Zhang Y, Teoh KH, Reed DW, Maes L, Goossens A, Olson DJH, Ross ARS,Covello PS. The molecular cloning of artemisinic aldehyde Delta 11(13)reductase and its role in glandular trichome-dependent biosynthesisof artemisinin in Artemisia annua. J Biol Chem 2008; 283: 21501–21508

23 Teoh KH, Polichuk DR, Reed DW, Covello PS. Molecular cloning of analdehyde dehydrogenase implicated in artemisinin biosynthesis inArtemisia annua. Botany 2009; 87: 635–642

24 Liu Y, Ye HC, Wang H, Li GF. Molecular cloning, Escherichia coli expres-sion and genomic organization of squalene synthase gene from Artemi-sia annua. Acta Bot Sin 2003; 45: 608–613

25 Yang RY, Feng LL, Yang XQ, Yin LL, Xu XL, Zeng QP.Quantitative transcriptprofiling reveals down-regulation of a sterol pathway relevant geneand overexpression of artemisinin biogenetic genes in transgenic Arte-misia annua plants. Planta Med 2008; 74: 1510–1516

26 Zhang L, Jing F, Li F, Li M,Wang Y, Wang G, Sun X, Tang K.Development oftransgenic Artemisia annua (Chinese wormwood) plants with an en-hanced content of artemisinin, an effective anti-malarial drug, by hair-pin-RNA-mediated gene silencing. Biotechnol Appl Biochem 2009; 52:199–207

1765

Chen J-L et al. Artemisinin Biosynthesis Enhancement… Planta Med 2011; 77: 1759–1765

Original Papers