PHYSIOLOGIA PLANTARUM 116: 79–86. 2002 Copyright C Physiologia Plantarum 2002

Printed in Denmark – all rights reserved ISSN 0031-9317

In vitro regeneration and genetic transformation of the berberine-producing plant, Thalictrum flavum ssp. glaucum

Nailish Samanani, Sang-Un Park and Peter J. Facchini*

Department of Biological Sciences, University of Calgary, Calgary, Alberta, T2N 1N4, Canada*Corresponding author, e-mail: pfacchin/ucalgary.ca

Received 13 February 2002

Protocols have been developed for the in vitro regenerationand Agrobacterium-mediated genetic transformation ofmeadow rue, Thalictrum flavum ssp. glaucum. Ten-day-oldseedlings were bisected along the embryonic axis and thecotyledons were co-cultured with various Agrobacterium tu-mefaciens strains for 3 days. The cotyledons were culturedon a shoot induction medium (B5 salts and vitamins, 30 glª1 sucrose, 2 mg lª1 kinetin, and 3 g lª1 Gelrite) contain-ing 25 mg lª1 hygromycin B as the selection agent and250 mg lª1 timentin to facilitate the elimination of Agrob-acterium. Only the oncogenic A.tumefaciens strains A281and C58 produced transgenic T.flavum callus tissues. A281was the most effective strain producing hygromycin-resistantcallus on 85% of the explants. Transgenic callus was subcul-

Introduction

The genus Thalictrum comprises more than 90 speciesand is one of many alkaloid-producing genera of theRanunculaceae, most of which are dispersed in northtemperate regions. Thalictrum species have long beenused for medicinal purposes in many parts of the world.For example, T. flavum was used in England, Russia, andNorth America in the eighteenth and nineteenth centur-ies as a laxative, diuretic, purgative, gastric stimulant,and fever suppressant (Schiff and Doskotch 1970). Thal-ictrum species accumulate the quaternary protoberberi-ne alkaloid berberine and the quaternary aporphine al-kaloid magnofluorine, which account for at least someof the pharmacological effects. Berberine stimulates thegastrointestinal tract, depresses auricles and ventricles,and dilates the heart, whereas magnofluorine induceshypotension and exhibits ganglionic blocking effects.

The biosynthesis of protoberberine alkaloids has beenstudied extensively. Several genes in the berberine bio-

Abbreviations – BA, 6-benzyladenine; B5, Gamborg et al. B5 medium; GA3, gibberellic acid; GUS, b-glucuronidase; HPT, hygromycin phospho-transferase.

Physiol. Plant. 116, 2002 79

tured on the shoot induction medium every 2 weeks. After12 weeks, hygromycin-resistant shoots that formed on ex-plants exposed to strain A281 were transferred to a rootinduction medium (B5 salts and vitamins, 25 mg lª1 hygro-mycin B, 250 mg lª1 timentin, and 3 g lª1 Gelrite). Detec-tion of the b-glucuronidase (GUS) gene using a polymerasechain reaction assay, the high levels of GUS mRNA andenzyme activity, and the cytohistochemical localization ofGUS activity confirmed the genetic transformation of calluscultures and regenerated plants. The transformation processdid not alter the normal content of berberine in transgenicroots or cell cultures; thus, the reported protocol is valuableto study the molecular and metabolic regulation of protob-erberine alkaloid biosynthesis.

synthetic pathway have been isolated from various mem-bers of the Ranunculaceae (Takeshita et al. 1995, Mori-shige et al. 2000, Choi et al. 2002). Although the basicmolecular mechanisms that regulate berberine biosyn-thesis are still not well understood, modern molecularand biochemical methodologies are expanding ourknowledge of the berberine pathway, and will createunique opportunities for its manipulation in transgenicplants. Genetic engineering strategies involving berber-ine biosynthesis in Thalictrum and related genera requirethe establishment of plant regeneration and transform-ation systems. Until now, methods for the genetic trans-formation of plants within the Ranunculaceae have notbeen reported. In this paper, we describe protocols forthe in vitro regeneration of meadow rue (T. flavum ssp.glaucum; also known as T. rugosum), a common peren-nial herb native to Europe and Asia, and the establish-ment of transgenic plants and cell cultures.

Materials and methods

Seed sterilization and germination

Seeds of T. flavum ssp. glaucum were purchased from Jel-itto Staudensamen (Schwarmstedt, Germany), andstored at 4æC. The seeds were surface-sterilized for 1min in 70% (v/v) ethanol, and for 10 min in 2% (v/v)sodium hypochlorite solution, then rinsed three timeswith sterilized water. Seeds were germinated in Petridishes on a medium containing B5 salts and vitamins(Gamborg et al. 1968), 30 mg lª1 sucrose, 3.0 mg lª1

GA3, and 3 g lª1 Gelrite (Schweizerhall, SouthPlainfield, USA). All media were adjusted to pH 5.8 andautoclaved at 1.1 kg cmª2 (121æC) for 20 min. Seedswere maintained at 4æC for 10 days, then transferred toa growth chamber. All tissues were maintained at 25æCunder standard cool white fluorescent tubes (SylvaniaGros-Lux Wide Spectrum, Mississauga, Canada) with a35-mmol sª1 mª2 flux rate and a 16-h photoperiod.

Plant regeneration procedure

Cotyledons from 10-day-old seedlings were isolated byremoving the hypocotyl below the cotyledonary node,longitudinally bisecting the embryonic axis, and cuttingoff the apical meristem. The cotyledons were transferredto a shoot induction medium (B5 salts and vitamins, 30g lª1 sucrose, 2 mg lª1 kinetin, and 3 g lª1 Gelrite) andsubcultured every 2 weeks. An effective concentration ofcytokinin in the shoot induction medium was deter-mined by testing different concentrations (1, 2, and 4 mglª1) of various cytokinins (BA, kinetin, and zeatin) onthe formation and growth of regenerated shoots. After8 weeks, regenerated shoots approximately 2 cm inlength were excised from the explant, and transferred toMagenta boxes (Sigma-Aldrich, St. Louis, MO, USA)containing 70 ml of a root induction medium (B5 saltsand vitamins and 3 g lª1 Gelrite). After 6 weeks, therooted plants were washed with sterile water to removethe Gelrite, transferred to pots containing autoclavedvermiculite, and covered with polyethylene bags for 1week to maintain high humidity. The plants were thentransferred to a soil mixture consisting of autoclavedbaked clay and peat at a ratio of 1:2, and maintained ina growth chamber with a photoperiod of 16 h and anight/day temperature of 18/20æC.

Preparation of Agrobacterium tumefaciens

The binary vector pWBVec10a was mobilized by elec-troporation in nine different A. tumefaciens strains:A281, AGL1, C58, EHA101, EHA105, GV2260,GV3101 containing the pMP90 helper plasmid (Konczand Schell 1986), GV3850, and LBA4404. ThepWBVec10a plasmid included a hygromycin phospho-transferase (HPT) gene, containing a castor bean cata-lase-1 (CAT1) gene intron, driven by the cauliflower mo-saic virus (CaMV) 35S promoter, and the b-glucuronida-se (GUS) gene driven by a maize ubiquitin (UBI1)

Physiol. Plant. 116, 200280

promoter, between T-DNA border sequences. The bi-nary vector pCAMBIA1305.1 (Cambia, Canberra, Aus-tralia) was mobilized in A. tumefaciens strain A281. Bac-terial cultures were grown to a density of A600 Ω 0.5 ona gyratory shaker at 180 g, 28æC in liquid Luria-Bertanimedium (1% [w/v] tryptone, 0.5% [w/v] yeast extract,and 1% [w/v] NaCl, pH 7.0), containing 50 mg lª1 specti-nomycin. Cells were collected by centrifugation at 1500g for 10 min, and resuspended at a density of A600 Ω 1.0in inoculation medium (B5 salts and vitamins and 30 glª1 sucrose).

Production of transgenic plants and cell cultures

Forty excised cotyledons bisected along the embryonicaxis were immersed for 15 min in suspension cultures ofthe various A. tumefaciens strains, blotted dry on sterilefilter paper and transferred to a co-cultivation medium(B5 salts and vitamins, 30 g lª1 sucrose, and 3 g lª1

Gelrite). After 3 days of co-cultivation with A. tumefaci-ens, the cotyledons were transferred to the shoot induc-tion medium containing 25 mg lª1 hygromycin B and250 mg lª1 timentin. Cotyledons and callus tissues thatformed were subcultured every 2 weeks on the same me-dium. After 8 weeks, regenerated shoots 1–2 cm inlength were excised from the explant tissue, and trans-ferred to the root induction medium containing 25 mglª1 hygromycin B and 250 mg lª1 timentin. After 6weeks, rooted plants were transferred to pots containingautoclaved vermiculite and then to soil.

Polymerase chain reaction analysis

Plant genomic DNA was extracted as described by Ed-wards et al. (1991). Fifty milligram of fresh tissue washomogenized in 200 ml of 0.5% [w/v] SDS, 250 mMNaCl, 100 mM Tris-HCl, pH 8.0, and 25 mM EDTA.The sample was centrifuged at 13 000 g and an equalvolume of isopropanol was added to the supernatant.The sample was then incubated on ice and centrifugedat 13 000 g. The pellet was dried and resuspended in100 ml 10 mM Tris-HCl, pH 7.4 and 1.0 mM EDTA.Polymerase chain reaction (PCR) was performed for 30thermal cycles (95æC for 1 min, 55æC for 1 min, and 72æCfor 1 min) using primers specific to the GUS reportergene (5ƒ-ATGTTACGTCCTGTAGAA-3ƒ and 5ƒ-TCATTGTTTGCCTCCCTG-3ƒ).

Assay of GUS activity

Tissues were extracted in 50 mM KPO4, pH 7.0, 1 MEDTA, and 10 m b-mercaptoethanol. 4-Methylumbel-liferyl-b--glucuronide (MUG; 0.44 mg mlª1) was addedto the assay buffer (50 mM NaPO4 buffer, pH 7.0, 10 mb-mercaptoethanol, 10 mM EDTA, 0.1% [w/v] sodiumlauryl sarcosine, and 0.1% [w/v] Triton X-100). Assayswere performed for 3 h at 37æC using 50 ml of tissueextract and 50 ml of assay buffer, and stopped with 500ml of 0.2M Na2CO3. 4-Methylumbelliferone (MU)

cleaved from MUG was quantified using a fluorescencespectrophotometer (Hitachi F-2000, Tokyo, Japan). Pro-tein concentrations were determined using the methodof Bradford (1976).

GUS histochemical staining

Histochemical staining for GUS activity was performedusing the method of Kosugi et al. (1990). Tissues werefixed in a 0.35% (v vª1) formaldehyde solution contain-ing 10 mM MES, pH 7.5, and 300 mM mannitol for 1 hat 20æC, rinsed three times in 50 mM sodium phosphate,pH 7.5, and subsequently incubated in 50 mM sodiumphosphate, pH 7.5, 10 mM EDTA, 300 mM mannitol,pH 7.0, and 1 mM 5-bromo-4-chloro-3-indolyl-b--gluc-uronide cyclohexylammonium salt for 6–12 h at 37æC.Stained tissues were rinsed extensively in 70% (v/v) etha-nol to remove residual chlorophyll.

RNA gel blot hybridization

Total RNA for gel blot hybridization analysis was iso-lated using the method of Logemann et al. (1987), and15 mg was fractionated on a 1.0% (v/v) formaldehydeagarose gel before transfer to a nylon membrane. Theblot was hybridized with a random-primer 32P-labelledfull-length GUS probe. Hybridization was performed at65æC in 0.25 mM sodium phosphate buffer, pH 8.0, 7%(w vª1) SDS, 1% (w vª1) BSA, and 1 mM EDTA. Theblot was washed at 65æC, twice with 2¿ SSC and 0.1%(w vª1) SDS and twice with 0.2¿ SSC and 0.1% (w/v)SDS (1¿ SSC Ω 0.15M NaCl and 0.015M sodiumcitrate, pH 7.0), and autoradiographed with an intensify-ing screen at ª80æC for 24 h.

HPLC analysis of protoberberine alkaloids

Protoberberine alkaloids were extracted in methanoland analysed by high pressure liquid chromatography(HPLC, System Gold 126, Beckman-Coulter, Mississau-ga, Canada) using a photodiode array detector (SystemGold 168, Beckman-Coulter). Alkaloids were separatedat a flow rate of 0.75 ml minª1 on a C18 reverse phasecolumn (4.6 ¿ 250 mm; Ultrasphere, Beckman-Coulter)using a gradient of methanol:water:acetic acid (60:40:1for 15 min, ramped to 90:10:1 over 5 min, maintainedat 90:10:1 for 10 min, ramped to 60:40:1 over 5 min, andmaintained at 60:40:1 for 5 min). The berberine peakwas identified by comparing its UV spectrum and reten-tion time to those of an authentic standard.

Results and discussion

Protocols have been developed for the in vitro regenera-tion and Agrobacterium-mediated genetic transform-ation of T. flavum ssp. glaucum. To establish a regenera-tion protocol useful for transformation purposes, we ini-tially investigated the effect of different concentrationsof various cytokinins on the efficiency of shoot organo-

Physiol. Plant. 116, 2002 81

genesis and growth in T. flavum (Table 1). Shoot develop-ment from cotyledonary explants did not occur in theabsence of exogenous cytokinin. Efficient developmentand growth of T. flavum shoots was promoted by the ad-dition of 2 mg lª1 kinetin or 1 mg lª1 zeatin (Table 1).In contrast, BA was less effective. Routinely, 2 mg lª1

kinetin was added to the shoot induction medium to in-duce shoot organogenesis.

Various stages of the T. flavum shoot organogenesisprocess are shown in Fig. 1. Less than 2 weeks after coty-ledonary explants were cultured on the shoot inductionmedium, the growth of callus tissue was visible at ex-cision and bisection sites (Fig. 1A). Numerous shootprimordia emerged from the callus within the next 3weeks (Fig. 1B). Small differentiated shoots developedfrom these primordia between 2 and 4 weeks later (Fig.1C,D). After 6 weeks, an average of 10–11 fully de-veloped shoots at least 1 cm in length were producedfrom each explant (Fig. 1E). Fully regenerated plantswere morphologically indistinguishable from seed-grownplants (Fig. 1F). This shoot organogenesis protocol forthe regeneration of T. flavum ssp. glaucum plants is simi-lar to those used for the transformation of opium poppy(Park and Facchini 2000c), alfalfa (Trieu and Harrison1996), and spinach (Zhang and Zeevaart 1999).

A variety of A. tumefaciens strains harbouring the vec-tor pWBVec10a were tested for their ability to transformT. flavum cotyledonary explants. Of the nine tested, onlythe transconjugant strain A281 and the wild-type strainC58 resulted in the formation of hygromycin-resistantcallus (Table 2). The frequency of callus formation usingstrain A281 was considerably higher than that of strainC58 (Table 2). A. tumefaciens A281 carrying plasmidpTiBo542 is a supervirulent strain that induces largerand faster growing tumours on a wider range of plantsthan most other strains, including C58 (Hood et al.1986). None of the disarmed A. tumefaciens strains

Table 1. Effect of different concentrations of cytokinins on the re-generation and growth of shoots from T. flavum ssp. glaucum seed-lings bisected along the embryonic axis and cultured for 8 weeks.

Number ofCytokinina Regeneration shoots Shoot lengthc

(mg lª1) frequencyb (%) per explantb (cm)

Control 0 – –1.0 7 4.0 ∫ 1.0 3.3 ∫ 1.32.0 40 3.5 ∫ 0.8 5.1 ∫ 2.94.0 13 2.0 ∫ 0.5 3.8 ∫ 1.2

Kinetin1.0 40 2.4 ∫ 0.9 4.1 ∫ 2.12.0 67 9.0 ∫ 5.2 5.1 ∫ 2.84.0 60 10.6 ∫ 6.5 7.1 ∫ 3.7

Zeatin1.0 73 6.8 ∫ 3.5 7.5 ∫ 4.92.0 70 5.4 ∫ 2.5 4.6 ∫ 2.34.0 80 3.7 ∫ 2.4 3.9 ∫ 2.3aBasal medium consisted of B5 salts and vitamins, 30 g lª1 sucrose,and 3 g lª1 Gelrite.

bFrom 15 cotyledons tested.cValues represent the mean ∫ of 25–50 shoots.

tested, including EHA101, were able to induce the for-mation of hygromycin-resistant tissues (Table 2). TheEHA101 strain has the same chromosomal backgroundas A281, but harbours a derivative of the pTiBo542plasmid from which the T-DNA region was removed(Hood et al. 1986). The supervirulence of strain A281on some members of the Solanaceae is encoded by aregion of pTiBo542 outside the T-DNA. However, ourdata indicate that the complete tumour-inducing (Ti)plasmid is essential for virulence on T. flavum.

Hygromycin-resistant callus that formed on T. flavumcotyledonary explants exposed to either A. tumefaciensstrain A281 or C58 was yellow to green in colour. Incontrast, the explants quickly became brown due to thepresence of hygromycin B in the medium. The presenceof the GUS gene in a variety of cultured callus lines wasdemonstrated by PCR using GUS-specific primers. Asingle amplicon with the expected size of 1809 bp wasobtained for all callus cultures tested. Five randomly se-

Fig. 1. Regeneration of T. flavum ssp.glaucum plants in vitro. (A) Initiation ofcallus formation on cotyledons excisedfrom seedlings bisected along theembryonic axis. (B) Numerous shootprimordia emerging after 3 weeks on theshoot induction medium. (C) Early shootdevelopment about 2 weeks after theappearance of primordia. (D) Shootdevelopment approximately 4 weeks afterthe appearance of primordia. (E) Fullydifferentiated shoots 1–2 cm in lengthready for transfer to the rooting mediumafter 8 weeks on the shoot inductionmedium. (F) Regenerated plant 8 weeksafter transfer to the rooting medium. Bar:(A) 2 mm; (B) 5 mm; (C) 5 mm; (D) 10mm; (E) 10 mm; and (F) 10 mm.

Physiol. Plant. 116, 200282

lected GUS-positive callus lines were tested for the pres-ence of GUS mRNAs and enzyme activity (Fig. 2). RNAgel blot hybridization analysis revealed abundant GUStranscript levels in each of the putative transgenic calluscultures (Fig. 2A). No signal was detected in a wild-typecontrol (Fig. 2A). Transgenic callus also showed highlevels of GUS activity compared to the wild-type culturewhich contained only a background level of GUS activ-ity (Fig. 2B). Although GUS mRNA levels were rela-tively similar in each transformed callus, GUS enzymeactivities varied about 2-fold from 1253 ∫ 64 to 2416 ∫196 pmol MU mgª1 protein minª1 (Fig. 2). Dispro-portional transgene mRNA and enzyme activity levelsare common in transgenic plants due to the effects oftransgene copy number, the location of chromosomal in-sertion, or various post-transcriptional events (Baul-combe 1996). Cytohistochemical staining for GUS activ-ity showed that most hygromycin-resistant callus cul-tures were transformed. The UBI1 promoter-GUS

Table 2. Effectiveness of different A. tumefaciens strains for the es-tablishment of transformed tissues from T. flavum ssp. glaucumseedlings bisected along the embryonic axis and cultured for 8weeks.

Frequency Frequencyof callus of transgenic plant

Agrobacterium formationa regenerationb

tumefaciens strain (%) (%)

Control 0 –A281 85 5c

AGL1 0 –C58 20 0d

EHA101 0 –EHA105 0 –GV2260 0 –GV3101-pMP90 0 –GV3850 0 –LBA4404 0 –aFrom 40 cotyledons tested.bFrequency reflects the number of transgenic plants relative to thetotal number of transformation events.

cFrom 35 independent calluses.dFrom 10 independent calluses.

cassette in pWBVec10a produced constitutive GUS ac-tivity (Fig. 3B) even though the promoter was of mono-cot origin. GUS-derived staining was not detected inwild-type callus (Fig. 3A). However, several hygromycin-resistant callus cultures composed of transgenic andwild-type cells were also obtained.

Although the frequency was low, hygromycin-resistantT. flavum plants could be regenerated from cotyledonaryexplants exposed to A. tumefaciens strain A281 (Table2). High levels of GUS mRNAs and enzyme activity,comparable to those found in callus, were also detectedin leaves of regenerated, hygromycin-resistant plants(data not shown). After root formation and transfer tosoil, the transgenic plants were morphologically indis-tinguishable from wild-type plants. Cytohistochemicalstaining indicated the presence of GUS activity in allcells of transgenic organs and tissues, whereas onlybackground levels of GUS activity were detected in wild-type plant materials (Fig. 3). However, we cannot ruleout the possibility that some cells in hygromycin-resis-tant tissues of primary transformants were not trans-genic. The preponderance of transgenic callus over re-generated plants was undoubtedly due to the presenceof the armed pTiBo542 plasmid in A. tumefaciens strainA281. It is possible that the tumour-inducing genes frompTiBo542 were either silenced or excluded from the T-DNA in the transgenic plants that were recovered. Therecovery of transgenic plants using the transconjugantstrain A281 has been reported for several species includ-ing Actinidia chinensis Planch. (Fraser et al. 1995) andMedicago sativa L. (Ninkovic et al. 1995). More typic-ally, only transgenic callus has been reported (Srinivasanand Sharma 1991, Sain et al. 1994). Clearly, the ef-ficiency of our transformation protocol could be im-proved if a disarmed Agrobacterium strain that displayedsupervirulence on T. flavum were identified. Althoughthe wild-type strain C58 has also been used for the pro-

Physiol. Plant. 116, 2002 83

duction of transgenic plants (Gartland et al. 2000, Shar-ma and Anjaiah 2000), no transgenic T. flavum plantswere obtained with this strain (Table 2).

The transformation of T. flavum was repeated usingA. tumefaciens strain A281 containing the binary vectorpCAMBIA1305.1, rather than pWBVec10a. The GUSgene in pCAMBIA1305.1 contains an intron; thus, thepresence of GUS transcript and enzyme activity in se-lectable marker-resistant tissues (data not shown) con-firms the expression of the reporter gene in plant cells.Moreover, A. tumefaciens strain A281 did not exhibitGUS activity with or without the binary vectors. Theseresults, combined with the absence of Agrobacterium intransformed tissues exhibiting ubiquitous GUS activity(Fig. 3), support the stable integration of the T-DNAinto the T. flavum genome.

Except for the Agrobacterium co-cultivation step, all

Fig. 2. (A) RNA gel blot hybridization analysis of the b-glucuronid-ase (GUS) reporter gene in wild-type (WT) and randomly selectedhygromycin-resistant (1–5) T. flavum ssp. glaucum callus cultures.The bands in lanes 1–5 correspond to the expected molecular weightof 1809 nucleotides for GUS transcripts. Fifteen mg of total RNAwas fractionated on a 1.0% formaldehyde agarose gel, transferredto a nylon membrane, and hybridized at high stringency with a 32P-labelled full-length probe for GUS (lower panel). The gel wasstained with ethidium bromide prior to blotting to ensure equalloading (top panel). (B) GUS activity in wild-type (WT) and hygro-mycin-resistant (1–5) T. flavum ssp. glaucum callus cultures using 4-methylumbelliferyl-b--glucuronide (MUG) as the substrate.

culture media included hygromycin B for the selectionof transformed plant tissues, and timentin to assist withthe elimination of the Agrobacterium after infection. Inpreliminary experiments, we examined the effects of hyg-romycin B on the viability of wild-type T. flavum coty-ledons. Hygromycin B caused necrosis of the cotyledonsin a dose-dependent manner at concentrations greaterthan 10 mg lª1. At a concentration of 25 mg lª1, hygro-mycin B, which is inactivated by the HPT gene product,was effective for the selection of transformed T. flavumtissues. T. flavum was relatively insensistive to kanamycinat concentrations up to 100 mg lª1. Although growthrates were reduced, wild-type shoots regenerated even atthe highest kanamycin concentrations tested. In con-trast, no wild-type plants escaped selection when hygro-mycin B was used. Insensitivity toward kanamycin andthe effectiveness of hygromycin B as a selective agenthave been reported for several other plant species (Wald-

Fig. 3. Histochemical localization of b-glucuronidase (GUS) activity inhygromycin-resistant T. flavum ssp.glaucum tissues transformed with theubiquitin (UBI1) promoter-GUS reportergene fusion. (A) Six-week-old wild-typeand (B) transgenic callus. (C) Wild-typeand (D) transgenic leaf of a regeneratedplant. (E) Wild-type and (F) transgenicroot of a regenerated plant. Alltransgenic tissues were derived fromcotyledonary explants exposed toAgrobacterium tumefaciens strain A281.

Physiol. Plant. 116, 200284

ron et al. 1985, Eady and Lister 1998, Kuvshinov et al.1999). Moreover, the CAT1 intron in the HPT codingregion of pWBVec10a completely abolished expressionof the gene in A. tumefaciens, rendering the bacteriumsusceptible to hygromycin B and minimizing its growthduring plant transformation (Wang et al. 1997). T. fla-vum cotyledons were also unaffected by timentin at con-centrations up to 400 mg lª1.

The quantity and profile of berberine and other proto-berberine alkaloids in wild-type and transgenic T. flavumroots and callus were essentially identical (Fig. 4). Themean berberine content of several transgenic and wild-type roots was 98 ∫ 26 and 85 ∫ 19 mg gª1 DW, respec-tively. Only trace amounts of protoberberine alkaloidswere detected in wild-type or transgenic shoot organs.We have also reported an unaltered benzylisoquinolinealkaloid profile for transgenic opium poppy (Park andFacchini 2000c) and California poppy plants transform-

Fig. 4. HPLC elution profiles of methanol extracts from wild-type(A) and transgenic (B) T. flavum ssp. glaucum roots 4 weeks afterthe transfer of plantlets to soil. Peaks identified with letters exhibitUV spectra with a protoberberine signature similar to that of ber-berine (peak c).

ed with the disarmed A. tumefaciens strain GV3101(Park and Facchini 2000b), and opium poppy and Cali-fornia poppy root cultures transformed with the armedA. rhizogenes strain R1000 (Park and Facchini 2000a).The unaltered profile of protoberberine alkaloids in T.flavum tissues transformed with A. tumefaciens strainA281 is significant because shoot cultures of opiumpoppy transformed with the oncogenic A. rhizogenesstrain MAFF 03-01724 were reported to accumulate analtered profile of the benzylisoquinoline alkaloids mor-phine and codeine (Yoshimatsu and Shimomura 1992).

The establishment of a protocol for the genetic trans-formation of T. flavum cell cultures and regeneratedplants is an important step in the development ofmethods for the metabolic engineering of protoberberi-

Physiol. Plant. 116, 2002 85

ne alkaloid biosynthesis. Recently, several genes involvedin the biosynthesis of berberine have been isolated fromCoptis japonica, a related member of the Ranunculaceae(Takeshita et al. 1995, Morishige et al. 2000, Choi et al.2002). Genes encoding conserved biosynthetic enzymesinvolved in the formation of related benzylisoquinolinealkaloids in members of the Papaveraceae have also beenreported (Facchini and De Luca 1994, Facchini et al.1996, Pauli and Kutchan 1998). The expanding collec-tion of cloned genes relevant to berberine biosynthesiscoupled with a protocol for the production of transgenicT. flavum ssp. glaucum plants and cell cultures providesunique opportunities to investigate the molecular regula-tion of protoberberine alkaloid biosynthesis, and to har-ness the biotechnological potential of metabolic engin-eering in medicinal members of the Ranunculaceae. Re-cently, cell cultures of C. japonica were transformed witha scoulerine O-methyltransferase gene and shown to ac-cumulate marginally higher levels of protoberberine al-kaloids (Sato et al. 2001). However, details of the trans-formation protocol were not provided.

Acknowledgements – We are grateful to the following individualsfor generously providing A. tumefaciens strains: Antonio Mercuri(A281), Alan Lloyd (GV3101-pMP90), Eugene Nester (C58 andA281), Dirk Inze (GV2260 and GV3850), and Doug Muench(AGL1). SUP was the recipient of Bettina Bahlsen Memorial,Graduate Faculty Council, and J.B. Hyne Graduate Scholarships,and a Dean’s Special Doctoral Scholarship, offered through theUniversity of Calgary. This research was funded by a NaturalSciences and Engineering Research Council of Canada grant to PJF.

ReferencesBaulcombe DC (1996) RNA as a target and initiator of post-tran-

scriptional gene silencing in transgenic plants. Plant Mol Biol32: 79–88

Bradford MM (1976) A rapid and sensitive method for the quant-itation of microgram quantities of protein utilizing the principleof protein-dye binding. Anal Biochem 72: 248–254

Choi K-B, Morishige T, Shitan N, Yuzaki K, Sato F (2002) Molecu-lar cloning and characterization of coclaurine N-methyltransfer-ase from cultured cells of Coptis japonica. J Biol Chem 277: 830–835

Eady CC, Lister CE (1998) A comparison of four selective agentsfor use with Allium cepa L. immature embryos and immatureembryo-derived cultures. Plant Cell Rep 18: 117–121

Edwards K, Johnstone C, Thompson C (1991) A simple and rapidmethod for preparation of plant genomic DNA for PCR analy-sis. Nucl Acids Res 19: 1349

Facchini PJ, De Luca V (1994) Differential and tissue-specific ex-pression of a gene family for tyrosine/dopa decarboxylase inopium poppy. J Biol Chem 269: 26684–26690

Facchini PJ, Penzes C, Johnson AG, Bull D (1996) Molecular char-acterization of berberine bridge enzyme genes from opiumpoppy. Plant Physiol 112: 1669–1677

Fraser LG, Kent J, Harvey CF (1995) Transformation studies ofActinidia chinensis Planch. NZ J Crop Hort Sci 23: 407–413

Gamborg OL, Miller RO, Ojima K (1968) Nutrient requirements ofsuspension cultures of soybean root cells. Exp Cell Res 50: 151–158

Gartland JS, McHugh AT, Brasier CM, Irvine RJ, Fenning TM,Gartland KM (2000) Regeneration of phenotypically normalEnglish elm (Ulmus procera) plantlets following transformationwith an Agrobacterium tumefaciens binary vector. Tree Physiol20: 901–907

Hood EE, Helmer GL, Fraley RT, Chilton MD (1986) The hyper-virulence of Agrobacterium tumefaciens A281 is encoded in a

region of pTiBo542 outside of T-DNA. J Bacteriol 168: 1291–1301

Koncz C, Schell J (1986) The promoter TL-DNA gene 5 controlsthe tissue-specific expression of chimeric genes carried by a noveltype of Agrobacterium vector. Mol General Genet 204: 383–396

Kosugi S, Ohashi Y, Nakajima K, Arai Y (1990) An improved assayfor b-glucuronidase in transformed cells: methanol almost com-pletely suppresses a putative endogenous b-glucuronidase activ-ity. Plant Sci 70: 133–140

Kuvshinov V, Koivu K, Kanerva A, Pehu E (1999) Agrobacterium-mediated transformation of greenhouse-grown Brassica rapa ssp.oleifera. Plant Cell Rep 18: 773–777

Logemann J, Schell J, Willmitzer L (1987) Improved method for theisolation of RNA from plant tissues. Anal Biochem 163: 16–20

Morishige T, Tsujita T, Yamada Y, Sato F (2000) Molecular char-acterization of the S-adenosyl--methionine: 3ƒ-hydroxy-N-methylcoclaurine-4ƒ-O-methyltransferase of isoquinoline alka-loid biosynthesis in Coptis japonica. J Biol Chem 275: 23398–23405

Ninkovic S, Miljus DJ, Neskovic M (1995) Genetic transformation ofalfalfa embryos and their clonal propagation through repetitivesomatic embryogenesis. Plant Cell Tiss Org Cult 42: 255–260

Park S-U, Facchini PJ (2000a) Agrobacterium rhizogenes-mediatedtransformation of opium poppy, Papaver somniferum L. & Cali-fornia poppy, Eschscholzia californica Cham. root cultures. JExp Bot 51: 1005–1016

Park S-U, Facchini PJ (2000b) Agrobacterium-mediated genetictransformation of California poppy, Eschscholzia californicaCham., via somatic embryogenesis. Plant Cell Rep 19: 1006–1012

Park S-U, Facchini PJ (2000c) Agrobacterium-mediated genetictransformation of opium poppy, Papaver somniferum L., viashoot organogenesis. J Plant Physiol 157: 207–214

Pauli HH, Kutchan TM (1998) Molecular cloning and functionalheterologous expression of two alleles encoding (S)-N-methylco-claurine 3ƒ-hydroxylase (CYP80B1), a new methyl jasmonate-inducible cytochrome P-450-dependent mono-oxygenase of ben-zylisoquinoline alkaloid biosynthesis. Plant J 13: 793–801

Edited by C. H. Bornman

Physiol. Plant. 116, 200286

Sain SL, Oduro KK, Furtek DB (1994) Genetic transformation ofcocoa leaf cells using Agrobacterium tumefaciens. Plant CellTissue Organ Cult 37: 243–251

Sato F, Hashimoto T, Hachiya A, Tamura K, Choi K-B, MorishigeT, Fulimoto H, Yamada Y (2001) Metabolic engineering ofplant alkaloid biosynthesis. Proc Natl Acad Sci USA 98: 367–372

Schiff PL, Doskotch RW (1970) Thalictrum alkaloids. Lloydia 33:403–452

Sharma KK, Anjaiah VV (2000) An efficient method for the pro-duction of transgenic plants of peanut (Arachis hypogaea L.)through Agrobacterium tumefaciens-mediated genetic trans-formation. Plant Sci 16: 7–19

Srinivasan Mohapatra T, Sharma RP (1991) Agrobacterium-me-diated genetic transformation of chickpea, Cicer arietinum L.Indian J Exp Biol 29: 758–761

Takeshita N, Fujiwara H, Mimura H, Fitchen JH, Yamada Y, SatoF (1995) Molecular cloning and characterization of S-adenosyl--methionine: scoulerine-9-O-methyltransferase from culturedcells of Coptis japonica. Plant Cell Physiol 36: 29–36

Trieu AT, Harrison MJ (1996) Rapid transformation of Medicagotruncatula: regeneration via shoot organogenesis. Plant Cell Rep16: 6–11

Waldron C, Murphy EB, Roberts JL, Gustafson GD, Armour SL,Malcolm SK (1985) Resistance to hygromycin B: a new markerfor plant transformation studies. Plant Mol Biol 5: 103–108

Wang M-B, Upadhyaya NM, Brettell RIS, Waterhouse PM (1997)Intron-mediated improvement of a selectable marker gene forplant transformation using Agrobacterium tumefaciens. J GenetBreed 51: 325–334

Yoshimatsu K, Shimomura K (1992) Transformation of opiumpoppy (Papaver somniferum L.) with Agrobacterium rhizogenesMAFF 03–01724. Plant Cell Rep 11: 132–136

Zhang H-X, Zeevaart JAD (1999) An efficient Agrobacterium tume-faciens-mediated transformation and regeneration system forcotyledons of spinach (Spinacia oleracea L.). Plant Cell Rep 18:640–645