Journal of Experimental Botany, Vol. 56, No. 420, pp. 26112618, October 2005 doi:10.

1093/jxb/eri253 Advance Access publication 16 August, 2005

RESEARCH PAPER

Enhanced secretion of tropane alkaloids in Nicotiana tabacum hairy roots expressing heterologous hyoscyamine-6b-hydroxylase
Suvi T. Hakkinen1,*, Elisabeth Moyano2, Rosa M. Cusido3, Javier Palazon3, M. Teresa Pinol3 and Kirsi-Marja Oksman-Caldentey1 VTT Biotechnology, Tietotie 2, PO Box 1500, FIN-02044 VTT (Espoo), Finland `ncies Experimentals i de la Salut, Universitat Pompeu Fabra, Avda. Dr Aiguader 80, Departament de Cie E-08003 Barcelona, Spain 3 n Seccio de Fisiologa Vegetal, Facultad de Farmacia, Universidad de Barcelona, Avda. Diagonal 643, E-08028 Barcelona, Spain
2 1

Downloaded from jxb.oxfordjournals.org by guest on October 27, 2010

Received 29 April 2005; Accepted 21 June 2005

Abstract
Hyoscyamine-6b-hydroxylase (H6H; EC 1.14.11.11) catalyses oxidative reactions in the biosynthetic pathway leading from hyoscyamine to the more pharmaceutically valuable tropane alkaloid scopolamine. The h6h gene encoding H6H from Hyoscyamus niger was introduced, under the control of the CaMV 35S promoter, into non-hyoscyamine-producing Nicotiana tabacum and hyoscyamine-producing Hyoscyamus muticus. The transformation was performed using a binary vector system based on Agrobacterium rhizogenes. Production of scopolamine in hairy roots was clearly correlated with the 35S-h6h transcript expression. The engineered N. tabacum and H. muticus hairy roots were studied for the production of scopolamine and other tropane and nicotine alkaloids after feeding the cultures with exogenous hyoscyamine. N. tabacum hairy roots carrying the 35S-h6h transgene showed a more efcient uptake of hyoscyamine from the culture medium and a higher rate of bioconversion of hyoscyamine to scopolamine than those of H. muticus. In particular, the secretion of scopolamine in N. tabacum hairy roots was remarkably high, up to 85% of the total scopolamine being released to the culture medium. Exogenous hyoscyamine also caused changes in nicotine alkaloid accumulation in N. tabacum hairy roots.
Key words: Hairy roots, hyoscyamine-6b-hydroxylase, Hyoscyamus, Nicotiana, secretion, tropane alkaloids.

Introduction Tropane alkaloids, for example, hyoscyamine and scopolamine, have pharmaceutical importance as anticholinergic agents, acting on the parasympathetic nervous system. Of the two, scopolamine is more valuable and is preferred for its higher physiological activity and fewer side-effects. Many plant-derived pharmaceuticals are still isolated from whole plants (e.g. Duboisia, Datura), this often being the only economically feasible production method. Plant cell cultures have been widely studied in order to obtain alternative production systems for compounds such as tropane alkaloids. The major bottleneck, however, is the complexity of the biosynthetic routes of these secondary compounds, involving unknown regulatory genes and enzymes (Oksman-Caldentey and Inze, 2004). Only a few functional genes of the tropane alkaloid pathway have been characterized and it is impossible to generalize about how the overexpression of genes involved in tropane alkaloid biosynthesis in different plant species might affect the end-product levels. Increased accumulation of the direct metabolite N-methylputrescine was observed when putrescine N-methyltransferase (pmt) was overexpressed in root cultures of Duboisia hybrids (Moyano et al., 2002), Atropa belladonna roots (Rothe et al., 2003), and Hyoscyamus muticus roots (Biondi et al., 2000), whereas the effect on the alkaloid level was only marginal (Moyano et al., 2003; Rothe et al., 2003). However, regulation of the expression of this gene has been shown to be crucial for alkaloid production in several genera, for example,

* To whom correspondence should be addressed. Fax: +358 20 722 7071. E-mail: Suvi.Hakkinen@vtt. The Author [2005]. Published by Oxford University Press [on behalf of the Society for Experimental Biology]. All rights reserved. For Permissions, please e-mail: journals.permissions@oupjournals.org

kkinen et al. 2612 Ha

Datura and Nicotiana (Robins et al., 1991, 1994; Pinol et al., 1999). Scopolamine is converted from hyoscyamine in a twostep process catalysed by the enzyme hyoscyamine-6bhydroxylase (H6H, EC 1.14.11.11). This enzyme catalyses both the hydroxylation of hyoscyamine leading to 6bhydroxyhyoscyamine and the epoxidation of the latter leading to scopolamine (Hashimoto and Yamada, 1986). It is localized in the pericycle of the root, where the tropane alkaloids are also synthesized and accumulated (Hashimoto et al., 1986; Matsuda et al., 1991). Furthermore, in studies with H6H isolated from A. belladonna, Suzuki et al. (1999) reported that AbH6H is expressed in root pericycle cells, in the tapetum, and in pollen mother cells. It has been shown that the scopolamine/hyoscyamine ratio can be increased in hairy roots of hyoscyamine-producing plants by overexpressing h6h (Hashimoto et al., 1993b). Yun et al. (1992) reported the successful introduction of the h6h gene from H. niger into a related species A. belladonna, resulting in almost complete conversion of hyoscyamine to scopolamine. Later, Jouhikainen et al. (1999) showed that overexpressing the same gene can also considerably enhance the scopolamine content in H. muticus hairy roots, while the hyoscyamine content remains unaltered. In these examples, the engineering of a single step in the pathway has led to an improved accumulation of the more valuable endproduct. Recently, Zhang et al. (2004) simultaneously overexpressed pmt and h6h genes in H. niger, resulting in a scopolamine production of 411 mg l1, which is the highest

level hitherto reported from cultivated hairy roots. Another interesting example of the expression of foreign genes in Solanaceous plants is the formation of resveratrol in Nicotiana as a result of heterologous expression of a gene from groundnut, coding for stilbene synthase (Hain et al., 1990). Nicotiana and Hyoscyamus, both belonging to the Solanaceae family, produce alkaloids from a common biosynthetic origin (Fig. 1). N-methylputrescine, which is converted from arginine/ornithine-derived putrescine, provides both the pyrroline moiety of nicotine and part of the tropane ring structure of hyoscyamine. Recently, Rocha et al. (2002) introduced both h6h and tropinone reductase I (trI) together into N. tabacum plants and followed the accumulation of alkaloids after feeding the detached leaves with hyoscyamine. In addition to the accumulation of the direct products of these enzymes, they also reported elevated nicotine alkaloid levels in these leaves. However, based on these results it is difcult to evaluate the effects of the individual genes alone, because they were introduced simultaneously. Furthermore, since nicotine and tropane alkaloids are mainly produced in the roots and further transported to the leaves of the intact plant, it would be necessary to examine the effects of the exogenously applied substrate in culture systems, where the actual biosynthesis of these alkaloids takes place. In this work, the potential of using Nicotiana tabacum hairy roots for the bioconversion of exogenously supplied hyoscyamine to more valuable scopolamine was assessed

Downloaded from jxb.oxfordjournals.org by guest on October 27, 2010

Fig. 1. Biosynthetic pathways of nicotine and tropane alkaloids. Abbreviations: LDC, lysine decarboxylase; PMT, putrescine N-methyltransferase; MPO, methylputrescine oxidase; TRI, tropinone reductase I; TRII, tropinone reductase II; H6H, hyoscyamine-6b-hydroxylase (note: one arrow may represent more than one reaction).

Secretion of tropane alkaloids in tobacco

2613

by overexpressing the h6h from H. niger in Nicotiana. Secretion of the alkaloids into the culture medium was studied and compared with that of the h6h overexpressing H. muticus hairy roots which naturally produce hyoscyamine. In addition, in this study the effect and function of h6h alone was more closely determined, and the overall alkaloid accumulation in both Nicotiana and Hyoscyamus hairy roots was investigated.

Materials and methods

Transformation of Nicotiana tabacum Leaves of Nicotiana tabacum cv. Xanthi plantlets grown in vitro on MS medium (Murashige and Skoog, 1962) were used for transformation with Agrobacterium rhizogenes LBA9402 wild type and A. rhizogenes LBA9402 pLAL21 carrying the 35S-h6h and the nptII gene as a marker, constructed by Jouhikainen et al. (1999). The hairy roots appeared 24 weeks after the infection and they were excised and cultured individually on MS medium supplemented with 30 g l1 sucrose and 500 ppm cefotaxime to eliminate the bacteria. For hairy roots carrying the 35S-h6h transgene, kanamycin (50 ppm) was used for selection. Hairy root clones were kept in the dark at 25 8C and routinely subcultured every 2530 d. After several subcultures, the root clones were transferred to MS liquid medium and kept on a rotary shaker at 100 rpm, 25 8C in the dark. Subculturing was performed every four weeks in fresh medium. Hairy roots of Hyoscyamus muticus The hairy roots of Hyoscyamus muticus (strain Cairo) were transformed with A. rhizogenes strains LBA9402 (control) and LBA9402 pLAL21 (carrying 35S-h6h) as described by Jouhikainen et al. (1999). The hairy roots were routinely grown and subcultured in modied B50 medium, as described by Oksman-Caldentey et al. (1991). The best scopolamine-producing clone KB7 (Jouhikainen et al., 1999) was chosen for feeding studies with hyoscyamine. PCR analysis Total DNA was isolated from hairy root clones according to Edwards et al. (1991). PCR analysis was performed using the preformulated, predispensed single dose reaction beads Ready to Go (Pharmacia Biotech). The complete mixture contained 200 ng of total DNA, 12.5 pmol ml1 of each oligonucleotide primer, 200 lM dNTPs, 1.5 ll Taq polymerase, and buffer supplied by the enzyme manufacturer (1/10V) in a total volume of 25 ll. The oligonucleotide primers used for amplication of the h6h gene were 59-CCG GAA TTC GGA TCC CAA CGT ATA GAT TCT TC-3 and 59-CGG GAA TTC GGA TCC CAA ACC ATC ACT GCA AT-39, according to the sequence of the h6h gene from Hyoscyamus niger (Matsuda et al., 1991), and produced a DNA fragment of 1153 bp. On the other hand, the primers used for amplication of the rolC gene (from A. rhizogenes T-DNA) were 59-TAA CAT GGC TGA AGA CGA CC-39 and 59-AAA CTT GCA CTC GCC ATG CC-39, according to the sequence of the gene described by Slightom et al. (1986), and produced a DNA fragment of 534 bp. PCR amplications of h6h and rolC were: initial denaturation at 95 8C for 5 min, followed by 30 cycles of denaturation at 94 8C for 1 min, annealing at 53 8C for 1 min, extension at 72 8C for 1.5 min, with a nal extension at 72 8C for 5 min. The PCR reaction mixtures (10 ll) were then loaded directly onto 1.5% agarose/TBE gel for electrophoretic analysis. A 100 Base-Pair Ladder (Pharmacia Biotech) was used as a molecular weight marker for the PCRamplied double-stranded DNA fragment.

Alkaloid analysis Nicotine alkaloids were extracted as described in Hakkinen et al. (2004) using 2,49-dipyridyl as an internal standard. Alkaloids were determined using a 6890 GC system tted with a 5973 Mass Selective Detector (MSD) (Agilent Technologies) operating at an ionization voltage of 70 eV (EI mode). The analyses were performed on an NB-54 fused silica capillary column (15 m, 0.20 mm i.d., HNU Nordion) by using a split sampling mode (50:1) and a programmed oven temperature from 70 8C to 100 8C (10 8C min1) and from 100 8C to 235 8C (15 8C min1). Injector and detector temperatures were kept at 250 8C. Aliquots of 2 ll were injected by a 7683 AutoSampler (Agilent Technologies). Identication of the nicotine alkaloids was performed based on the mass fragmentation and the retention order (Hakkinen et al., 2004). The tropane alkaloids scopolamine and hyoscyamine were extracted and analysed from N. tabacum cultured hairy roots and culture medium with the method described by Plank and Wagner (1986), and alkaloids from H. muticus hairy roots as follows. Lyophilized roots were weighed (50 mg) and the lipids were removed with 2 ml petroleum ether. After vortexing and centrifugation (3000 rpm, 10 min) the solvent phase was discarded and the sample residue was dried under nitrogen ow. After adding 50 lg internal standard homatropine (Sigma), the samples were made alkaline (pH 9) and the alkaloids were extracted twice with Cl2CH2. Before GC-MS analysis, the dried residue was dissolved in 40 ll Cl2CH2 and derivatized with MSTFA (N-methylN-triuoroacetamide, Pierce). The tropane alkaloids from the culture medium were extracted correspondingly from 2 ml medium. The GCMS conditions were as desribed in Goossens et al. (2003a) and alkaloids and alkaloid derivatives were identied based on the GCMS spectral data (Hartmann et al., 1986; Witte et al., 1987). Northern blot analysis Extraction of total RNA from N. tabacum hairy roots after 2 weeks of culture in MS medium was performed using the RNeasy kit (Qiagen). Ten micrograms of total RNA were loaded per lane of a denaturating formaldehyde gel (Sambrook et al., 1989). Northern blots were made on Hybond-N+ nylon transfer membrane (Amersham) and hybridized with 32P-labelled probe specic for h6h (full cDNA). Hybridization was performed at 42 8C in the presence of 50% formamide. The blots were washed at 65 8C with saline sodium citrate buffer (SSC), 0.1% SDS (30 min) and with 13 SSC, 0.1% SDS (15 min). The radioactivity on the lter was imaged using Hyperlm-MP (Amersham). Feeding assays Hairy root cultures of N. tabacum and H. muticus were fed with hyoscyamine added to the culture medium. The roots, weighing 50065 mg and 10065 mg from N. tabacum and H. muticus, respectively, were grown in 40 ml liquid MS and 20 ml B50 medium, respectively. The hyoscyamine hydrochloride (Sigma) was dissolved in water and the stock was added to a nal concentration of 100 mg l1 or 200 mg l1. In control samples, only water was added. The hairy roots were grown as described above. Alkaloid levels were determined both in the roots and in the culture medium after a culturing period of 28 d.

Downloaded from jxb.oxfordjournals.org by guest on October 27, 2010

Results and discussion

Characterization and capacity for hyoscyamine bioconversion of N. tabacum transgenic hairy root clones

Ten (10) hairy root clones carrying the 35S-h6h transgene (named H6H) and three (3) control clones, showing good

kkinen et al. 2614 Ha

growth and typical hairy root morphology were analysed for their transformed nature. All the H6H clones gave two bands of 1.15 and 0.53 kb after PCR amplication, the former corresponding to the h6h fragment and the latter to the rolC fragment, whereas the control clones only gave the band corresponding to the rolC gene (data not shown). The hyoscyamine uptake percentage and the biomass production of the control hairy root clones and the clones carrying the 35S-h6h transgene are shown in Table 1. The uptake of the added hyoscyamine was calculated from the amount of hyoscyamine measured from the culture medium at day 28. The degradation of hyoscyamine over 28 d in the culture conditions was found to be negligible. The uptake was, on average, 95% (8499%) with 100 mg l1 hyoscyamine, and 94% (78100%) with 200 mg l1 hyoscyamine. The biomass accumulation measured after 28 d varied between 0.14 and 0.51 g DW in 40 ml MS medium. The addition of 100 mg l1 hyoscyamine had no signicant effect on the growth, but hyoscyamine added at a higher concentration (200 mg l1) caused a statistically signicant reduction in the growth of all clones (ANOVA F2,36=18.532, P <0.001, for normalized values). Hyoscyamine concentrations of 300 mg l1 and above had clear toxic effects on N. tabacum hairy roots and were therefore not included in the studies. In order to establish a possible relationship between the 35S-h6h transgene expression and the capacity of the transgenic hairy root clones to produce scopolamine, i.e. to carry out the bioconversion of exogenously applied hyoscyamine, the expression levels of the 35S-h6h transgene in the H6H clones were assayed by northern blot analysis. As shown in Fig. 2, the transgene expression was observed in H6H 1, H6H 14, and H6H 101, which were the only clones to produce scopolamine. This correlation

is in accordance with earlier results that conrm h6h gene expression to be essential for scopolamine production in plant systems (Hashimoto and Yamada, 1986; Jouhikainen et al., 1999). The bioconversion to scopolamine was highest when 100 mg l1 hyoscyamine was used, the best clones H6H 1 and H6H 14 having conversion rates of 45% and 40%, respectively (Table 2). The total (medium and root-associated) scopolamine content was again highest in the clones H6H 1 and H6H 14 (Table 2). When the amount of added hyoscyamine was doubled (from 100 mg l1 to 200 mg l1), the production of scopolamine increased by 11% and 53% (calculated

Downloaded from jxb.oxfordjournals.org by guest on October 27, 2010

Fig. 2. Northern blot results corresponding to the one wild-type (LBA1) and the ten (10) transgenic hairy root clones of N. tabacum. The probe was full cDNA for the H. niger h6h gene. Each clone was charged with the same quantity of total RNA. (+) Positive control A. rhizogenes LBA9402pLAL21.

Table 2. Hyoscyamine (root associated) and total scopolamine (medium and root associated) contents in N. tabacum hairy root cultures (LBA, control; H6H transgenic) at the end of the culture period (28 d) in MS liquid medium supplemented with 100 mg l1 or 200 mg l1 hyoscyamine
Each value is the average of three replicates 6SD. Hyoscyamine Scopolamine Scopolamine Scopolamine total (mg l1) secretion (%) conversiona (mg l1) (%) MS + Hyoscyamine 100 mg l1 LBA 1 LBA 2 LBA 3 H6H 1 H6H 14 H6H 101 90.060.9 91.961.3 80.061.7 49.160.6 48.160.6 87.662.0 ndb nd nd 47.361.0 41.960.4 11.060.3 83 56 40 45 40 10

Table 1. Dry weight (g) of control (LBA) and transgenic (H6H) hairy root clones of N. tabacum at the end of the culture period (28 d) in 40 ml MS liquid medium with or without exogenous supply of hyoscyamine (Hyos)
The corresponding hyoscyamine uptake (%) is given in brackets. Each value is the average of three replicates 6SD. MS LBA 1 LBA 2 LBA 3 H6H 1 H6H 2 H6H 10 H6H 14 H6H 17 H6H 26 H6H 38 H6H 101 H6H 105 H6H 115 0.47360.029 0.41060.038 0.47660.040 0.29160.018 0.50960.048 0.36260.033 0.47460.047 0.27360.033 0.38060.029 0.39360.029 0.25260.031 0.49660.046 0.13960.028 MS+Hyos 100 mg l1 0.45660.022 0.36360.017 0.46660.036 0.28060.033 0.48860.029 0.35460.041 0.44860.039 0.25160.017 0.36660.033 0.25660.031 0.23360.026 0.47160.041 0.11060.009 (91) (93) (84) (98) (98) (99) (95) (95) (98) (93) (98) (99) (94) MS+Hyos 200 mg l1 0.42360.035 0.33960.029 0.46060.017 0.21160.023 0.48460.019 0.27560.046 0.37760.032 0.23660.038 0.31060.041 0.22660.022 0.10960.016 0.34460.033 0.08960.029 (89) (94) (91) (100) (98) (99) (98) (99) (98) (92) (78) (97) (95)

MS+Hyoscyamine 200 mg l1 LBA 1 LBA 2 LBA 3 H6H 1 H6H 14 H6H 101

a b

175.567.9 187.266.8 180.269.9 147.065.8 129.265.9 123.564.6

nd nd nd 52.563.2 64.163.7 5.760.1

85 62 0

25 31 3

Conversion, hyoscyamine to scopolamine bioconversion (molar ratio). nd, Not detected.

Secretion of tropane alkaloids in tobacco

2615

from Table 2) in the root clones H6H 1 and H6H 14, respectively. Thus, the most efcient absolute conversion rate was obtained with 100 mg l1 hyoscyamine. The overall scopolamine production was high, even up to 64.1 mg l1 (clone H6H 14). The low amounts of scopolamine produced by clone H6H 101 were most probably linked to the weak h6h expression (Fig. 2).
Production of tropane alkaloids in transgenic H. muticus hairy roots after feeding with hyoscyamine

Exogenous hyoscyamine was effectively taken up by both control and KB7 (with the 35S-h6h transgene) clones; the residual hyoscyamine found in the medium was less than 3% and less than 7% when hyoscyamine was added in concentrations of 100 mg l1 and 200 mg l1, respectively (Table 3). However, by contrast with the result observed for hairy root cultures of N. tabacum, exogenous hyoscyamine addition in concentrations even up to 500 mg l1 did not affect the growth of H. muticus hairy roots (data not shown). Regarding the alkaloid production, although the root-associated hyoscyamine content in the control clone (without the 35S-h6h transgene) increased linearly with an increasing amount of added substrate, the control clone produced only trace amounts of total (medium and root associated) 6b-hydroxyhyoscyamine and no scopolamine 28 d after feeding (Table 3). Compared with non-fed clones, the concentration of total hyoscyamine found in fed control clones was 3-fold (with 200 mg l1). The total hyoscyamine content in hyoscyamine-fed roots was less than expected (135 mg l1 and 218 mg l1 after feeding 100 mg l1 and 200 mg l1, respectively), taking into account the endogenous production of 77 mg l1 (Table 3). There are several possibilities causing the lower hyoscyamine content than expected, for instance feedback inhibition of certain biosynthetic steps by high concentration of this alkaloid, or degradation of the added precursor that can be used as nutrients by the hairy roots (Hashimoto and Yamada, 1983). In clone KB7 (carrying the 35S-h6h transgene) the endogenous hyoscyamine production was higher than that

of the control clone (Table 3). The addition of hyoscyamine resulted in a higher production of the intermediate 6b-hydroxyhyoscyamine, whereas the accumulation of scopolamine increased only slightly with 100 mg l1 hyoscyamine (Table 3). Earlier, it was shown that, in the system in which hyoscyamine is fed, the H6H activity becomes limiting and causes the accumulation of 6bhydroxyhyoscyamine (Hashimoto et al., 1993a). When 100 mg l1 hyoscyamine was added to the cultures, the conversion rate to scopolamine was 15% (Table 3). However, since the uptake of hyoscyamine did not increase linearly with the increasing concentrations of exogenous hyoscyamine, the conversion rate to scopolamine was clearly lower in 200 mg l1 feeding, being only 2% (Table 3). In another study, Hashimoto and Yamada (1983) reported approximately 20% conversion from hyoscyamine to scopolamine in adventitious roots of H. niger after feeding with 0.1 mM (29 mg l1) hyoscyamine. However, with 1 mM (290 mg l1) hyoscyamine the conversion decreased to around 10%. Compared with the control clone, the KB7 clone also produced slightly higher amounts of other minor alkaloids, for example pseudotropine, apoatropine, and 6b-hydroxyapoatropine (data not shown). The latter was not detected in the control clone. The amounts of apoatropine and 6bhydroxy-apoatropine increased with the increasing concentrations of added hyoscyamine, whereas the accumulation of pseudotropine remained stable.
Secretion of tropane alkaloids In N. tabacum hairy root clones carrying the 35S-h6h transgene, scopolamine was efciently secreted extracellularly; up to 85% of the produced scopolamine was detected in the culture medium (Table 2). This secretion rate is high compared with other reported values, when only up to 20% of the produced scopolamine has been found in the culture medium of Hyoscyamus or Atropa sp. (Hashimoto et al., 1986, 1993b; Jouhikainen et al., 1999). However, after feeding with hyoscyamine (1 mM), Hashimoto and Yamada (1983) reported that approximately 6070% of the total

Downloaded from jxb.oxfordjournals.org by guest on October 27, 2010

Table 3. Total (medium and root associated) hyoscyamine (Hyos), 6b-hydroxyhyoscyamine (6BHH) and scopolamine (Scop) content, and respective accounts (%) in the medium in control and 35Sh6h carrying (KB7) H. muticus hairy roots at the end of the culture period (28 d) in modied B50 liquid medium supplemented with 100 mg l1 or 200 mg l1 hyoscyamine
Each value is the average of three replicates 6SD. Hyos (mg l1) Control Control+100 mg l1 Hyos Control+200 mg l1 Hyos KB7 KB7+100 mg l1 Hyos KB7+200 mg l1 Hyos
a b

6BHH (mg l1) 260 160 360 3862 6663 7364

Scop (mg l1) ndb nd nd 3762 5263 4062

Hyos in medium (%) 0.8 1.5 3.0 5.5 2.1 6.7

6BHH secretion (%) 0.0 0.0 0.0 3.5 3.6 9.4

Scop secretion (%) 0.0 0.0 0.0 7.8 6.8 11.9

Scop conversiona (%) 15 2

7765 13567 21869 12466 185613 203610

Conversion, hyoscyamine to scopolamine bioconversion (molar ratio). nd, Not detected.

kkinen et al. 2616 Ha

scopolamine produced was detected in the extracellular medium. It was found that, compared with transgenic N. tabacum hairy root clones, in H. muticus carrying the 35S-h6h transgene, the scopolamine production was lower and, in particular, the secretion of scopolamine into the medium (712%) decreased considerably (Table 3). It is known that some secondary compounds, even though produced by the plant itself, may cause intrinsic toxicity to the cells. Scopolamine is a foreign substance for N. tabacum, in contrast to Hyoscyamus in which scopolamine is a compound endogenously produced. It may become toxic to N. tabacum cells in high concentrations, and is therefore efciently transported to the surrounding medium. Recently, it has been shown with N. tabacum cell suspension cultures, that the secretion of nicotine alkaloids can be stimulated and the tolerance to exogenously supplied alkaloids (hyoscyamine and scopolamine) can be enhanced by using PDR-type ATP binding cassette transporters (Goossens et al., 2003b). Considering large-scale production systems, extracellular secretion of the product is desirable, as it reduces the costs during both production and downstream processing, allowing the use of a continuous culturing system and direct recovery of the metabolites from the medium.
Nicotine alkaloids Nicotine alkaloids were determined from two control clones (LBA 1 and LBA 2) and two 35S-h6h transgenic clones (H6H 14 and H6H 101), exhibiting the highest and the lowest capacities for scopolamine production, respectively. N. tabacum hairy root clones produced nicotine as the major alkaloid, followed by anatabine, anatalline, anabasine, and nornicotine. The total root-associated nicotine alkaloid content varied in the range 0.635.10 mg g1 DW and 2.175.43 mg g1 DW in control and transgenic clones, respectively (Table 4). Interestingly, when hyoscyamine was fed to the cultures, the nicotine alkaloid levels were altered. The total alkaloid content in the control hairy root clones was increased 26-fold, mainly contributing to

the increment in nicotine levels. It is doubtful whether hyoscyamine could be used as a substrate for nicotine biosynthesis. More probably, the resulting increased biosynthetic activity could be caused by recognition of hyoscyamine as a foreign substance, thus causing an elicitory effect in N. tabacum. Several nicotine alkaloids in Nicotiana have previously been shown to be induced by elicitors, for example, methyl jasmonate (Imanishi et al., 1998; Goossens et al., 2003a). In the clones carrying the 35-h6h transgene, the exposure to hyoscyamine did not cause such a remarkable increase in nicotine alkaloid contents, and clone H6H 101 even showed lower levels of all alkaloids with a hyoscyamine concentration of 200 mg l1 than that observed in the case of the non-fed clone (Table 4). Since the transgenic clones had the capacity of converting hyoscyamine into scopolamine, the activity of the secondary metabolite biosynthesis was directed towards the tropane alkaloid branch putatively at the expense of endogenous nicotine alkaloid production. Earlier, Rocha et al. (2002) have also reported 313-fold higher concentrations of nicotine in the leaves of h6h- and tropinone reductase I-transformed N. tabacum clones. It was considered that these changes in the nicotine metabolism arose from perturbation of the normal biosynthesis due to the genetic engineering of the biosynthetic pathway. When the nicotine alkaloid levels of seven individual 35S-h6h transformed root clones were analysed, a slight increment in alkaloid levels of transgenic clones compared with the alkaloid levels of the control clones was observed, mainly accounting for the increase in nicotine and nornicotine (Table 5). However, taking into account the high variation in the total nicotine alkaloid levels in individual clones (0.65.1 mg g1 DW in control clones and 0.3 6.8 mg g1 DW in h6h-transformed clones), this increase was not considered signicant. In addition, no correlation between the alkaloid levels and H6H expression (Fig. 2) was observed. Therefore, these observations support the role of exogenously applied foreign phytoalexin, hyoscyamine, rather than that of the transformation of the 35S-h6h gene alone. The variation in the alkaloid production

Downloaded from jxb.oxfordjournals.org by guest on October 27, 2010

Table 4. The nicotine alkaloid contents of N. tabacum wild-type (LBA 1 and LBA 2) and transgenic hairy root clones (H6H 14 and H6H 101, carrying the 35Sh6h transgene) at the end of the culture period (28 d) in MS liquid medium, and supplemented with 200 mg l1 hyoscyamine (Hyos)
Each value is the average of two replicates6SD. Nicotine (mg g1 DW) LBA 1 LBA 1+200 mg l1 Hyos LBA 2 LBA 2+200 mg l1 Hyos H6H 14 H6H 14+200 mg l1 Hyos H6H 101 H6H 101+200 mg l1 Hyos 0.4460.09 2.9860.04 3.2060.25 7.3160.17 2.9660.98 5.1360.53 1.5260.51 0.2060.05 Nornicotine (mg g1 DW) 0.0260.00 0.0460.00 0.1060.02 0.1960.00 0.1760.02 0.1160.00 0.0460.00 0.0060.00 Anabasine (mg g1 DW) 0.0260.00 0.0960.00 0.1260.01 0.1660.00 0.1660.01 0.1560.01 0.0860.00 0.0060.00 Anatabine (mg g1 DW) 0.1160.00 0.4860.01 0.8660.03 0.8260.00 1.2260.07 0.9960.03 0.3360.00 0.0260.00 Anatalline (mg g1 DW) 0.0560.00 0.4260.03 0.8160.02 0.7860.01 0.9260.02 0.4160.04 0.1860.03 0.0260.00 Total (mg g1 DW) 0.63 4.01 5.10 9.26 5.43 6.79 2.17 0.24

Secretion of tropane alkaloids in tobacco

2617

Table 5. The nicotine alkaloid contents of N. tabacum wild type (LBA 1 and LBA 2) and seven transgenic hairy root clones carrying the 35Sh6h transgene at the end of the culture period (28 d) in MS liquid medium
Each value is the average of two replicates 6SD. Nicotine (mg g1 DW) LBA 1 LBA 2 Mean LBA (n=2) H6H 1 H6H 14 H6H 17 H6H 26 H6H 38 H6H 101 H6H 105 Mean H6H (n=7) 0.4460.09 3.2060.25 1.82 2.4160.01 2.9660.02 3.1160.00 4.9760.00 1.9260.00 1.5260.00 0.2760.00 2.45 Nornicotine (mg g1 DW) 0.0260.00 0.1060.02 0.06 0.0560.00 0.1760.01 0.1660.00 0.1560.00 0.2160.01 0.0460.00 0.0060.00 0.11 Anabasine (mg g1 DW) 0.0260.00 0.1260.01 0.07 0.0760.00 0.1660.01 0.1860.00 0.1760.00 0.1560.01 0.0860.00 0.0060.00 0.11 Anatabine (mg g1 DW) 0.1160.00 0.8660.03 0.49 0.4060.00 1.2260.07 1.2560.01 0.8960.01 1.2760.00 0.3360.00 0.0360.00 0.77 Anatalline (mg g1 DW) 0.0560.00 0.8160.02 0.43 0.3260.03 0.9260.02 0.9960.04 0.6560.03 1.2560.12 0.1860.03 0.0260.00 0.62 Total (mg g1 DW) 0.63 5.10 2.87 3.24 5.43 5.68 6.83 4.79 2.17 0.33 4.07

capacity of H. muticus hairy roots has been extensively studied before (Sevon et al., 1998). Although the long-term hyoscyamine production of root clones was shown to be stable for up to 50 passages of subculturing, the production between individual protoplast-derived hairy root clones varied up to 40-fold, between 0.04% and 1.55% DW (n=171). Recently, Palazon et al. (2003) obtained hairy roots of Duboisia overexpressing the h6h gene from H. niger and reported elevated total alkaloid contents for this strain, which endogenously produces both hyoscyamine and scopolamine. In these roots, the feedback inhibition caused by the accumulated hyoscyamine diminished, thus resulting in elevated overall alkaloid levels. In this study it was shown that hyoscyamine was efciently taken up by the N. tabacum roots, and it is unclear whether hyoscyamine could result in lowered feedback inhibition in nicotine alkaloid biosynthesis, for example, affecting intracellular signalling mechanisms, and further lead to elevated nicotine alkaloid accumulation in hyoscyamine-fed roots. No nicotine alkaloids were observed in H. muticus hairy roots, either in control or hyoscyamine-fed roots. This was an expected result, as there are no previous reports indicating nicotine alkaloid biosynthesis in Hyoscyamus sp. Hitherto, only some Solanaceae plants such as Duboisia sp. as well as Anthocercis, Lycopersicon, and Solanum sp. are known to be capable of producing both tropane and nicotine alkaloids (Evans and Ramsey, 1979; Siegmund et al., 1999). Conclusion It has been shown that the hairy root cultures of Nicotiana tabacum expressing the h6h gene from H. niger have a high capacity for converting exogenously applied hyoscyamine into pharmaceutically valuable scopolamine. Furthermore, the scopolamine was efciently secreted to the medium. Exogenous application of the foreign substrate hyoscyamine

resulted in enhanced production of various nicotine alkaloids, suggesting that the regulation of the production of these defence-related compounds is probably more complex than presently known. The mechanism by which hyoscyamine functions at the cellular level requires further studies, for example, whether exogenous hyoscyamine affects the expression levels of individual genes or enzymes, or whether it can trigger the overall up-regulation of the whole metabolic pathway. The use of transformed hairy root cultures offers the great advantages of combining metabolic engineering approaches in systems that are genetically and biochemically stable and which possess the capacity for high secondary metabolite production. Thus, the application of fast-growing plant cell culture systems, such as hairy roots, will offer new insights and possibilities in improving the production efciencies of known or novel high-value compounds of plant origin.

Downloaded from jxb.oxfordjournals.org by guest on October 27, 2010

Acknowledgements
The authors thank Dr Tuulikki Seppanen-Laakso and Dr Into Laakso for help in GC-MS analysis and Kari Kammiovirta for skilful technical assistance. STH is a recipient of a predoctoral fellowship of the Finnish Graduate School on Applied Bioscience. This work was supported by a grant from the Spanish MCYT (BIO2002-03614 and BIO2002-02328) and from the National Technology Agency of Finland (Tekes) programme NeoBio to KMOC.

References
Biondi S, Fornale S, Oksman-Caldentey K-M, Eeva M, Agostani S, Bagni N. 2000. Jasmonates induce over-accumulation of methylputrescine and conjugated polyamines in Hyoscyamus muticus L. root cultures. Plant Cell Reports 19, 691697. Edwards E, Johnstone C, Thompson A. 1991. A simple and rapid method for the preparation of plant genomic DNA for PCR analysis. Nucleic Acids Research 19, 1349. Evans WC, Ramsey KPA. 1979. Alkaloids of Anthocercis frodosa (Solanaceae). Journal of Pharmacy and Pharmacology 31, 9.

kkinen et al. 2618 Ha

Goossens A, Hakkinen ST, Laakso I, et al. 2003a. A functional genomics approach toward the understanding of secondary metabolism in plant cells. Proceedings of the National Academy of Sciences, USA 100, 85958600. Goossens A, Hakkinen ST, Laakso I, Oksman-Caldentey K-M, Inze D. 2003b. Secretion of secondary metabolites by ATPbinding cassette transporters in plant cell suspension cultures. Plant Physiology 131, 11611164. Hain R, Bieseler B, Kindl H, Schroder G, Stocker R. 1990. Expression of a stilbene synthase in Nicotiana tabacum results in synthesis of phytoalexin resveratrol. Plant Molecular Biology 15, 325335. Hartmann T, Witte L, Oprach F, Toppel G. 1986. Reinvestigation of the alkaloid composition of Atropa belladonna plants, root cultures, and cell suspension cultures. Planta Medica 52, 390395. Hashimoto T, Yamada Y. 1983. Scopolamine production in suspension cultures and redifferentiated roots of Hyoscyamus niger. Planta Medica 47, 121141. Hashimoto T, Yamada Y. 1986. Hyoscyamine 6b-hydroxylase, a 2-oxoglutarate-dependent dioxygenase, in alkaloid-producing root cultures. Plant Physiology 81, 619625. Hashimoto T, Yukimune Y, Yamada Y. 1986. Tropane alkaloid production in Hyoscyamus root cultures. Journal of Plant Physiology 124, 6175. Hashimoto T, Matsuda Y, Yamada Y. 1993a. Two-step epoxidation of hyoscyamine to scopolamine is catalysed by bifunctional hyoscyamine 6-hydroxylase. FEBS Letters 329, 3539. Hashimoto T, Yun D-J, Yamada Y. 1993b. Production of tropane alkaloids in genetically engineered root cultures. Phytochemistry 32, 713718. Hakkinen ST, Rischer H, Laakso I, Maaheimo H, SeppanenLaakso T, Oksman-Caldentey K-M. 2004. Anatalline and other methyl jasmonate-inducible nicotine alkaloids from Nicotiana tabacum cv. BY-2 cell cultures. Planta Medica 70, 936941. Imanishi S, Hashizume K, Nakakita M, Kojima H, Matsubayashi Y, Hashimoto T, Sakagami Y, Yamada Y, Nakamura K. 1998. Differential induction by methyl jasmonate of genes encoding ornithine decarboxylase and other enzymes involved in nicotine biosynthesis in tobacco cell cultures. Plant Molecular Biology 38, 11011111. Jouhikainen K, Lindgren L, Jokelainen T, Hiltunen R, Teeri TH, Oksman-Caldentey K-M. 1999. Enhancement of scopolamine production in Hyoscyamus muticus L. hairy root cultures by genetic engineering. Planta 208, 545551. Matsuda J, Okabe S, Hashimoto T, Yamada Y. 1991. Molecular cloning of hyoscyamine 6b-hydroxylase, a 2-oxoglutaratedependent dioxygenase, from cultured roots of Hyoscyamus niger. Journal of Biological Chemistry 266, 94609464. Murashige T, Skoog F. 1962. A revised medium for rapid growth and bioassays with tobacco tissue cultures. Physiologia Plantarum 15, 473497. Moyano E, Fornale S, Palazon J, Cusido RM, Bagni N, Pinol MT. 2002. Alkaloid production in Duboisia hybrid hairy root cultures overexpressing the pmt gene. Phytochemistry 59, 697702. Moyano E, Jouhikainen K, Tammela P, Palazon J, Cusido RM, Pinol MT, Teeri T, Oksman-Caldentey K-M. 2003. Effect of pmt gene overexpression on tropane alkaloid production in transformed root cultues of Datura metel and Hyoscyamus muticus. Journal of Experimental Botany 54, 203211. Oksman-Caldentey K-M, Inze D. 2004. Plant cell factories in the post-genomic era: new ways to produce designer secondary metabolites. Trends in Plant Science 9, 433440. Oksman-Caldentey K-M, Kivela O, Hiltunen R. 1991. Spontaneus shoot organogenesis and plant regeneration from hairy root cultures of Hyoscyamus muticus. Plant Science 78, 129136. Palazon J, Moyano E, Cusido RM, Bonll M, Oksman Caldentey K-M, Pinol MT. 2003. Alkaloid production in Duboisia hybrid hairy roots and plants overexpressing the h6h gene. Plant Science 165, 12891295. Pinol MT, Palazon J, Cusido RM, Ribo M. 1999. Inuence of calcium ion-concentration in the medium on tropane alkaloid accumulation in Datura stramonium hairy roots. Plant Science 141, 4149. Plank KH, Wagner KG. 1986. Determination of hyoscyamine and scopolamine in Datura innoxia plants by high performance liquid chromatography. Zeitung fu Naturforschung 41, 391395. r Robins RJ, Bent EG, Rhodes MJC. 1991. Studies on the biosynthesis of tropane alkaloids by Datura stramonium L. transformed root cultures. Planta 185, 385390. Robins RJ, Walton NJ, Parr AJ, Aird ELH, Rhodes MJC, Hamill JD. 1994. Progress in the genetic engineering of the pyridine and tropane alkaloid biosynthetic pathways of Solanaceous plants. In: Ellis BE, Kuroki GW, Stafford HA, eds. Genetic engineering of plant secondary metabolism. New York: Plenum Press, 133. Rocha P, Stenzel O, Parr A, Walton N, Christou P, Drager B, Leech MJ. 2002. Functional expression of tropinone reductase I (trI) and hyoscyamine-6b-hydroxylase (h6h) from Hyoscyamus niger in Nicotiana tabacum. Plant Science 162, 905913. Rothe G, Hachiya A, Yamada Y, Hashimoto T, Drager B. 2003. Alkaloids in plants and root cultures of Atropa belladonna overexpressing putrescine N-methyltransferase. Journal of Experimental Botany 54, 20652070. Sambrook J, Fritsch EF, Maniatis T. 1989. Molecular cloning: a laboratory manual. Cold Spring Harbor, New York: Cold Spring Harbor Laboratory Press. Sevon N, Hiltunen R, Oksman-Caldentey K-M. 1998. Somaclonal variation in transformed roots and protoplast-derived hairy root clones of Hyoscyamus muticus. Planta Medica 64, 3741. Siegmund B, Leitner E, Pfannhauser W. 1999. Determination of the nicotine content of various edible nightshades (Solanaceae) and their products and estimation of the associated dietary nicotine intake. Journal of Agricultural and Food Chemistry 47, 31133120. Slightom JL, Duran-Tardif M, Jouanin L, Tepfer D. 1986. Nucleotide sequence analysis of TL-DNA of Agrobacterium rhizogenes agropine type plasmid: identication of open-reading frames. Journal of Biological Chemistry 261, 108121. Suzuki K, Yun D-J, Chen X-Y, Yamada Y, Hashimoto T. 1999. An Atropa belladonna hyoscyamine 6b-hydroxylase gene is differentially expressed in the root pericycle and anthers. Plant Molecular Biology 40, 141152. Witte L, Muller K, Arfmann H-A. 1987. Investigation of the alkaloid pattern of Datura innoxia plants by capillary gasliquid-chromatography-mass-spectrometry. Planta Medica 53, 192197. Yun DJ, Hashimoto T, Yamada Y. 1992. Metabolic engineering of medicinal plants: transgenic Atropa belladonna with an improved alkaloid composition. Proceedings of the National Academy of Sciences, USA 89, 1179911803. Zhang L, Ding R, Chai Y, et al. 2004. Engineering tropane biosynthetic pathway in Hyoscyamus niger hairy root cultures. Proceedings of the National Academy of Sciences, USA 101, 67866791.

Downloaded from jxb.oxfordjournals.org by guest on October 27, 2010