Integrating carbon-halogen bond formation into medicinal plant metabolism

Article abstract of DOI:10.1038/nature09524

Halogenation, which was once considered a rare occurrence in nature, has now been observed in many natural product biosynthetic pathways. However, only a small fraction of halogenated compounds have been isolated from terrestrial plants. Given the impact that halogenation can have on the biological activity of natural products, we reasoned that the introduction of halides into medicinal plant metabolism would provide the opportunity to rationally bioengineer a broad variety of novel plant products with altered, and perhaps improved, pharmacological properties. Here we report that chlorination biosynthetic machinery from soil bacteria can be successfully introduced into the medicinal plant Catharanthus roseus (Madagascar periwinkle). These prokaryotic halogenases function within the context of the plant cell to generate chlorinated tryptophan, which is then shuttled into monoterpene indole alkaloid metabolism to yield chlorinated alkaloids. A new functional group-a halide-is thereby introduced into the complex metabolism of C. roseus, and is incorporated in a predictable and regioselective manner onto the plant alkaloid products. Medicinal plants, despite their genetic and developmental complexity, therefore seem to be a viable platform for synthetic biology efforts.

Full text of DOI:10.1038/nature09524

LETTER
doi:10.1038/nature09524
Integrating carbon–halogen bond formation into
medicinal plant metabolism
Weerawat Runguphan1*, Xudong Qu1{* & Sarah E. O’Connor1
Halogenation, which was once considered a rare occurrence in nat- C. roseus, and is incorporated in a predictable and regioselective
ure, has now been observed in many natural product biosynthetic manner onto the plant alkaloid products. Medicinal plants, despite
pathways1. However, only a small fraction of halogenated com- their genetic and developmental complexity, therefore seem to be a
pounds have been isolated from terrestrial plants2. Given the impact
viable platform for synthetic biology efforts.
that halogenation can have on the biological activity of natural pro-
Numerous halogenase enzymes from soil bacteria have been iden-
ducts1, we reasoned that the introduction of halides into medicinal tified and characterized extensively1,3–5. Two of these flavoenzymes,
plant metabolism would provide the opportunity to rationally PyrH6,7 and RebH8–11, chlorinate the indole ring of tryptophan in the
bioengineer a broad variety of novel plant products with altered, five and seven positions, respectively. Transferring these enzymes into
and perhaps improved, pharmacological properties. Here we report other natural product pathways would allow site-specific incorpora-
that chlorination biosynthetic machinery from soil bacteria can be tion of halogens onto a range of tryptophan-derived alkaloid pro-
successfully introduced into the medicinal plant Catharanthus ducts12, provided that the downstream enzymes could accommodate
the chlorinated tryptophan precursor.
Catharanthus roseus produces a wide variety of monoterpene indole
roseus (Madagascar periwinkle). These prokaryotic halogenases
function within the context of the plant cell to generate chlorinated
tryptophan, whichisthenshuttled into monoterpeneindolealkaloid alkaloids13 (Fig. 1a). This metabolic pathway begins with the conver-
metabolism toyieldchlorinatedalkaloids. A newfunctional group— sion of tryptophan (1) to tryptamine (2) by tryptophan decarboxy-
a halide—is thereby introduced into the complex metabolism of lase14. Tryptamine then condenses with the iridoid terpene
a
N
N
H
CO2H
NH2
Strictosidine
synthase
NH2
NH
H
H
Tryptophan
decarboxylase
O-Glc
N
CHO
N
N
CO2Me
H
H
5
7
H
CO2Me
N
N
O-Glc
O
H
H
Tryptophan (1)
Tryptamine (2)
O
MeO2C
N
MeO2C
N
H
O
Strictosidine (4)
Secologanin (3)
N
N
CO2Me
H
H
8
H
6
MeO2C
N
CO2H
NH2
Tryptophan
decarboxylase
(C. roseus)
NH2
b
3
RebH
reductase
H
CO2Me
N
N
H
N
H
H
X
X
X
CO2H
NH2
X = Cl: 5a
X = Br: 5c
X = Cl: 7-chlorotryptamine (2a)
X = Br: 7-bromotryptamine (2c)
4
5
X = Cl: 7-chlorotryptophan (1a)
X = Br: 7-bromotryptophan (1c)
N
H
Cl
6
N
CO2H
O
N
Tryptophan
decarboxylase
(C. roseus)
H
7
STRvm
3
H
NH2
H
Cl
6b
NH2
Tryptophan (1)
Cl
MeO2C
N
H
PyrH
reductase
Cl
N
N
H
H
5-chlorotryptophan (1b)
5-chlorotryptamine (2b)
N
H
CO2Me
7b
Figure 1
|
Monoterpene indole alkaloid biosynthesis. a, Tryptophan (1) is
pharmacological activities24–26,30. Me, CH3; Glc, glucose. b, RebH and PyrH,
alongwitha partner reductase, halogenate the indolering oftryptophanto yield
chlorotryptophan. Here we show that after transformation of these enzymes
into C. roseus, the halogenated tryptophans 1a and 1b can be decarboxylated by
tryptophan decarboxylase (C. roseus) to form the chlorotryptamines 2a and
decarboxylated by tryptophan decarboxylase to yield tryptamine (2), which
reacts with secologanin (3) to form strictosidine (4). After numerous
rearrangements, strictosidine (4) is converted into a variety of monoterpene
indole alkaloids, such as 19,20-dihydroakuammicine (5), ajmalicine
(6), tabersonine (7) and catharanthine (8). These compounds have a variety of 2b and then converted into chlorinated monoterpene indole alkaloids.
1Department of Chemistry, Massachusetts Institute of Technology, Cambridge, Massachusetts 02139, USA. {Present address: State Key Laboratory of Bioorganic and Natural Products Chemistry,
Shanghai Institute of Organic Chemistry, Chinese Academy of Sciences, 345 Lingling Road, Shanghai 200032, China.
*These authors contributed equally to this work.
1 8 N O V E M B E R 2 0 1 0
| V O L 4 6 8 | N A T U R E | 4 6 1
Macmillan Publishers Limited. All rights reserved
©2010

RESEARCH LETTER
secologanin (3) to form a biosynthetic intermediate strictosidine (4), 0.099 mM21 min21), 7-chlorotryptophan (1a; Km 5 499 6 74 mM,
which is subsequently functionalized in C. roseus to form over 100
k
cat 5 1.6 6 0.04 min21, kcat/Km 5 0.00327 mM21 min21) and 5-chlor-
alkaloids, including the anticancer agent vinblastine13. Previous work otryptophan (1b; Km 5 538 6 48 mM, kcat 5 2.56 0.08 min21, kcat
/
has shown that when C. roseus cell culture is supplemented with a Km 5 0.00455 mM21 min21) (Supplementary Figs 1 and 2). The activity
variety of halogenated tryptamines, the corresponding halogenated of the enzyme suggested that halogenated tryptophan could be decar-
alkaloid analogues are produced in isolable yields15,16. If prokaryotic boxylated in vivo.
halogenases could function in the eukaryotic plant cell, and if trypto-
phan decarboxylase could convert halogenated tryptophan into halo-
genated tryptamine, then C. roseus would produce chlorinated
alkaloids de novo (Fig. 1b).
When considering how to merge the prokaryotic biosynthetic
machinery with the plant alkaloid pathway, we chose to transfer the
halogenase enzymes into C. roseus rather than move the plant biosyn-
thetic enzymes into a microbial host. Most of the monoterpene indole
Because RebH and PyrH do not turn over tryptamine, this strategy alkaloid biosynthetic genes have not been identified, making hetero-
requires that tryptophan decarboxylase from C. roseus recognize halo- logous expression of this pathway impossible at present. Moreover, we
genated tryptophan. We assayed tryptophan decarboxylase from C. note that reconstitution of plant alkaloid pathways continues to be a
roseus in vitro with tryptophan (Km 5 51.7 6 9.2 mM (Michaelis challenge17,18. Many alkaloids use complex starting materials (such as
constant),
k
cat 5 5.16 0.1 min21 (turnover number), kcat/Km 5 secologanin) that are only produced by a few specialized plants, so
a
b
10-chloro-
ajmalicine (6b)
standard
12-chloro-19,20-dihydro-
akuammicine (5a) standard
m/z 387
m/z 359
m/z 359
Wild-type line
Wild-type line
m/z 387
RebF–RebH
m/z 359
PyrH–RebF–
STRvm
m/z 387
1.8 2.0 2.2 2.4 2.6 2.8
3.2
2.0
2.2
2.4
H10
2.6
2.8
3.0
3.2
1.6
3.0
Time (min)
Time (min)
1H NMR
H9
c
d
H11
1H–13C HSQC
H11
110
120
130
140
10-chloro-
15-chloro-
catharanthine (8b)
tabersonine (7b)
m/z 371
m/z 371
H10
5a
H9
6.5
8.0 7.5 7.0
1H f1 (p.p.m.)
7.5 7.4 7.3 7.2 7.1 7.0 6.9
H10
Wild-type line
H9
H11
110
120
130
140
H11
5c
PyrH–RebF–
STRvm
m/z 371
H10
H9
2.0
2.2
2.4
2.6
2.8
3.0
3.2
6.5
7.5 7.4 7.3 7.2 7.1 7.0 6.9
8.0
7.5
7.0
1H f1 (p.p.m.)
Time (min)
f1 (p.p.m.)
Figure 2
|
Chlorinated alkaloids in C. roseus hairy root culture. a, LC–MS showing 15-chlorotabersonine (7b) in RebF–PyrH–STRvm hairy roots (blue
chromatograms showing 12-chloro-19,20-dihydroakuammicine (5a; m/z 359) trace), contrasted with control cultures (purple trace). An authentic standard of
in RebF–RebH hairy roots (red trace), contrasted with control cultures
transformed with no plasmid (purple trace). An authentic standard of
7b is shown22 (black trace). The other major peak at m/z 371 had an exact mass
and ultraviolet spectrum consistent with a chlorinated analogue of
5a validated the structural assignment (black trace; Supplementary Figs 30 and catharanthine22 (8) (Supplementary Fig. 29). d, 1H NMR and 1H–13C
31). b, Chromatograms showing 10-chloroajmalicine (6b) in RebF–PyrH–
heteronuclear single quantum coherence (HSQC) spectra of 5a and 5c. f1 and
STRvm hairy roots (blue trace), contrasted with control cultures (purple trace). f2, chemical shifts in the 1H and 13C dimensions, respectively.
An authentic standard of 6b is shown22 (black trace). c, Chromatograms
4 6 2
| N A T U R E | V O L 4 6 8 | 1 8 N O V E M B E R 2 0 1 0
Macmillan Publishers Limited. All rights reserved
©2010

LETTER RESEARCH
reconstitution of plant alkaloid pathways must also include biosyn- 12 and 14). Different concentrations of KCl (3 mM–20 mM) were
thesis of these precursors. For example, ajmalicine (6; Fig. 1a), one of added to the medium, but increasing amounts of exogenous chloride
the simplest of the monoterpene indole alkaloids, requires an esti- salt did not significantly affect the yields of chlorinated alkaloids
mated 14 discrete enzymes for biosynthesis from tryptophan and the (Supplementary Figs 17 and 18).
terpene geraniol13; reconstitution of a pathway of this length is a sig-
nificant engineering problem. Therefore, we believe that exploring
Previous reports demonstrated that RebH can use bromide to yield
brominated tryptophan8 (1c). To assess the capacity of RebH for bro-
approaches in the host plant is an important aspect of alkaloid meta-
bolic engineering efforts.
mination in vivo, we supplemented a low-chloride cell culture medium
with KBr. The in vitro halide specificity of RebH correlated with the
products generated in vivo, as we observed the formation of a com-
pound that co-eluted with an authentic standard of 12-bromo-19,20-
dihydroakuammicine (5c) (21 6 8 and 49 6 20 mg per gram of fresh
root weight with 10 mM and 20 mM KBr supplementation, respect-
ively; Fig. 3). In contrast, supplementation of the medium with KI
failed to yield either iodinated tryptophan or iodinated alkaloids.
Again, this correlated with in vitro studies showing that RebH does
not accept iodide as a substrate8 (Supplementary Figs 19–22).
We also measured the transcript levels of the heterologous enzymes
by real-time PCR with reverse transcription. The production of halo-
genated compounds depended on the expression of both RebF and
RebH or PyrH. Notably, when the strictosidine synthase mutant
STRvm was not expressed in the PyrH–RebF hairy root lines, we
observed accumulationof 5-chlorotryptophan(representativecell line,
9 6 1 mg per gram of fresh root weight) and 5-chlorotryptamine (rep-
resentative cell line, 20 6 9 mg per gram of fresh root weight), but did
not observe downstream alkaloids (Supplementary Figs 23 and 24).
Tryptophan does not seem to accumulate in either wild-type or
transformed hairy roots. However, accumulation of 7-chlorotrypto-
phan (50 6 12 mg per gram of fresh root weight for a representative
RebH–RebF cell line) and 5-chlorotryptophan (8 6 2 mg per gram of
fresh root weight for a representative PyrH–RebF–STRvm cell line)
was observed, suggesting that decarboxylation of chlorinated trypto-
phan is a bottleneck in vivo, a step thatcould potentially be subjected to
future engineering efforts. This is consistent with the 30-fold-lower
catalytic efficiency of the decarboxylase enzyme for halogenated
To produce 7-chlorotryptophan in planta, we generated an express-
ion construct containing codon-optimized complementary DNA
encoding the 7-tryptophan chlorinase RebH and its required partner
flavin reductase, RebF, in a plant expression vector (pCAMBIA1300),
both under the control of constitutive cauliflower mosaic virus
(CaMV) 35S promoters. For production of 5-chlorotryptophan, an
expression construct encoding the 5-chlorinating enzyme PyrH, along
with RebF as the partner reductase, was generated. No signal sequence
was added to the halogenase genes, to ensure that RebH, PyrH and
RebF would produce chlorinated tryptophan in the cytosol, where it
would most readily encounter the decarboxylase, which is also loca-
lized in the cytosol19 (Supplementary Figs 3–5).
We used Agrobacterium rhizogenes to generate hairy root culture of
C. roseus transformed with the halogenase genes20. One of the early
biosynthetic enzymes, strictosidinesynthase, cannotturnover5-chlor-
otryptamine21 (2b). Therefore, when transforming C. roseus with pyrH
and rebF, we also introduced a mutant of strictosidine synthase
(STRvm) that can convert 5-chlorotryptamine to 10-chlorostrictosi-
dine16,22 (4b). After a selection process, we cultivated the transformed
root culture on standard Gamborg’s B5 plant medium and monitored
chlorinated alkaloids using liquid chromatography/mass spectrometry
(LC–MS). We observed formation of chlorinated tryptophans 1a and
1b and chlorinated alkaloidsin boththe RebH–RebF andPyrH–RebF–
STRvm hairy root lines (Fig. 2 and Supplementary Figs 6–15). These
results indicate that RebH, PyrH and the partner reductase function
productively in the plant cell environment, demonstrating that the
flavin halogenases are highly transportable among kingdoms.
Because chlorinated alkaloid production was observed in the trans-
formed lines, we conclude that tryptophan decarboxylase can compe-
tently turn over halogenated tryptophan substrates in vivo.
Hairy roots transformed with RebH and RebF, which produce
7-chlorotryptophan, yielded a major chlorinated product at m/z 359
(Fig. 2a). An authentic standard of 12-chloro-19,20-dihydroakuam-
micine (5a) co-eluted with this compound. Natural products contain-
ing the akuammicine scaffold have a variety of pharmacological
activities23–25. Although the parent compound, 19,20-dihydroakuam-
micine (5) has been isolated in good yields from other plants26, it is not
a major alkaloid in C. roseus hairy root culture. However, when wild-
type C. roseus cell lines were incubated with 7-chlorotryptamine, 12-
chloro-19,20-dihydroakuammicine was also the major chlorinated
product (Supplementary Fig. 16). Therefore, the predominance of
12-chloro-19,20-dihydroakuammicine in the RebH–RebF hairy root
line is probably due to substrate specificity of downstream enzymes for
7-chlorotryptamine. A hairy root line transformed with the 5-chloro-
tryptophan enzyme system, PyrH, RebF and STRvm, produced a
variety of chlorinated alkaloids (Fig. 2b–d). Two representative chlori-
nated alkaloids, 10-chloroajmalicine (6b) and 15-chlorotabersonine
(7b), were identified by co-elution with authentic standards21.
Chlorinated alkaloid production seemed to be stable over the course
of at least six subcultures. The alkaloid 12-chloro-19,20-dihydroa-
kuammicine was produced at 26 6 4 mg per gram of fresh root weight
of a representative cell line averaged over six subcultures. For compar-
ison, wild-type cell lines produced ,25 mg per gram of fresh tissue
weight of chlorinated alkaloids when the medium was supplemented
with 200 mM 7-chlorotryptamine (2a). Similarly, 10-chloroajmalicine
and 15-chlorotabersonine (7b) were produced at 2.8 6 0.9 and
4.0 6 1.0 mg per gram of fresh root weight, respectively, for a repres-
12-bromo-19,20-dihydro-
akuammicine (5c) standard
m/z 403
m/z 403
Wild-type line
+ 20 mM KBr
Wild-type line
+ 10 mM KBr
m/z 403
m/z 403
m/z 403
m/z 403
RebF–RebH
+ 20 mM KBr
RebF–RebH
+ 10 mM KBr
RebF–RebH
+ 0 mM KBr
2.0
2.2
2.4
2.6
2.8
3.0
3.2
Time (min)
Figure 3
|
Extracted LC–MS chromatograms showing the presence of 12-
bromo-19,20-dihydroakuammicine (5c; m/z 403) in RebF–RebH hairy
roots. Hairy roots are grown in medium supplemented with KBr (0–20 mM
final concentration) for two weeks before alkaloid extractions. 12-bromo-
19,20-dihydroakuammicine is not observed in control cultures transformed
with no plasmid after incubation in KBr-supplemented medium. An authentic
standard of 12-bromo-19,20-dihydroakuammicine is used to validate the
entative cell line averaged over four subcultures (Supplementary Figs structural assignment (Supplementary Figs 29, 30 and 32).
1 8 N O V E M B E R 2 0 1 0
| V O L 4 6 8 | N A T U R E | 4 6 3
Macmillan Publishers Limited. All rights reserved
©2010

RESEARCH LETTER
´
4. van Pee, K. H. & Patallo, E. P. Flavin-dependent halogenases involved in secondary
tryptophan in vitro. The morphologies of the halogen producing
lines were thicker and slower growing than those of wild-type lines
(Supplementary Fig. 25). Because tryptophan serves as the precursor
for other small-molecule metabolites, we speculate that chlorinated
tryptophan may be diverted into other pathways such as auxins.
Notably, 4-chloro indole acetic acid, which is found in several species
metabolism in bacteria. Appl. Microbiol. Biotechnol. 70, 631–641 (2006).
5. Blasiak, L. C. & Drennan, C. L. Structural perspective on enzymatic halogenation.
Acc. Chem. Res. 42, 147–155 (2009).
6. Zehner, S. et al. A regioselective tryptophan 5-halogenase is involved in
pyrroindomycin biosynthesis in Streptomyces rugosporus LL-42D005. Chem. Biol.
12, 445–452 (2005).
7. Zhu, X. etal.Structural insightsintoregioselectivity inthe enzymatic chlorination of
tryptophan. J. Mol. Biol. 391, 74–85 (2009).
of pea, has altered activity relative to the auxin indole acetic acid27,28
.
8. Yeh, E., Garneau, S. & Walsh, C. T. Robust in vitro activity of RebF and RebH, a two-
component reductase/halogenase, generating 7-chlorotryptophan during
rebeccamycin biosynthesis. Proc. Natl Acad. Sci. USA 102, 3960–3965 (2005).
9. Yeh, E. et al. Flavin redox chemistry precedes substrate chlorination during the
reaction of the flavin-dependent halogenase RebH. Biochemistry 45, 7904–7912
(2006).
10. Yeh, E., Blasiak, L. C., Koglin, A., Drennan, C. L. & Walsh, C. T. Chlorination by a long-
lived intermediate in the mechanism of flavin-dependent halogenases.
Biochemistry 46, 1284–1292 (2007).
Medicinal plants produce a wide range of complex natural products
but generate relatively few halogenated compounds; chlorinated or
brominated compounds are not found among the approximately
3,000 known monoterpene indole alkaloids produced by plants in
the Apocynaceae, Rubiaceae and Loganiaceae families. The halogena-
tion of natural products often has profound effects on the bioactivity of
the compound, and can serve as a useful handle for further chemical
derivatization1,29. Despite the metabolic and developmental complex-
11. Bitto, E. et al. The structure of flavin-dependent tryptophan 7-halogenase RebH.
Proteins 70, 289–293 (2008).
´
ity of plant tissue, transformation of these prokaryotic genes led to the 12. Sanchez, C. et al. Combinatorial biosynthesis of antitumor indolocarbazole
compounds. Proc. Natl Acad. Sci. USA 102, 461–466 (2005).
regioselective incorporation of halides into the alkaloid products of the
existing plant pathway. Notably, the yield of chlorinated alkaloids in
the most productive lines (,26 mg per gram of fresh weight of plant
tissue) is only 15-fold lower than the yield of total natural alkaloids
13. O’Connor, S. E. & Maresh, J. Chemistry and biology of terpene indole alkaloid
biosynthesis. Nat. Prod. Rep. 23, 532–547 (2006).
14. De Luca, V., Marineau, C. & Brisson, N. Molecular cloning and analysis of cDNA
encoding a plant tryptophan decarboxylase: comparison with animal dopa
decarboxylases. Proc. Natl Acad. Sci. USA 86, 2582–2586 (1989).
15. McCoy, E. & O’Connor, S. E. Directed biosynthesis of alkaloid analogs in the
(compounds 5, 6, 7 and 8) from wild-type tissue (,420 mg per gram of
fresh weight of plant tissue) (Supplementary Fig. 26). The ease with
which we engineered the successful production of chlorinated alka-
loids in C. roseus, a plant with limited genetic characterization, indi-
cates that medicinal plants can provide a viable platform for synthetic
biology.
medicinalplantCatharanthusroseus.J.Am. Chem. Soc.128,14276–14277(2006).
16. Bernhardt, P., McCoy, E. & O’Connor, S. E. Rapid identification of enzyme variants
for reengineered alkaloid biosynthesis in periwinkle. Chem. Biol. 14, 888–897
(2007).
17. Hawkins, K. M. & Smolke, C. D. Production of benzylisoquinoline alkaloids in
Saccharomyces cerevisiae. Nature Chem. Biol. 4, 564–573 (2008).
18. Minami, H. et al. Microbial production of plant benzylisoquinoline alkaloids.
Proc. Natl Acad. Sci. USA 105, 7393–7398 (2008).
19. De Luca, V. & Cutler, A. J. Subcellular localization of enzymes involved in indole
alkaloid biosynthesis in Catharanthus roseus. Plant Physiol. 85, 1099–1102 (1987).
20. Hughes, E. H., Hong, S.-B., Shanks, J. V., San, K.-Y. & Gibson, S. I. Characterization of
an inducible promoter system in Catharanthus roseus hairy roots. Biotechnol. Prog.
18, 1183–1186 (2002).
21. Runguphan, W. & O’Connor, S. E. Metabolic reprogramming of periwinkle plant
culture. Nature Chem. Biol. 5, 151–153 (2009).
METHODS SUMMARY
Structural characterization is shown in Supplementary Figs 27–32 and
Supplementary Tables 1 and 2.
Generation of transgenic C. roseus hairy root cultures. We transformed the
expression construct pCAMRebHRebF into A. rhizogenes ATCC 15834 by elec-
troporation (1-mm cuvette, 1.25 kV), and we co-transformed pCAMPyrHRebF
and pCAMSTRvm into A. rhizogenes ATCC 15834 by electroporation.
Transformation of C. roseus seedlings with the generated Agrobacterium strains
was performed as previously reported20.
Evaluation of alkaloid production in transgenic C. roseus hairy roots. Every
transgenic hairy root line that survived hygromycin selection medium was evalu-
atedforalkaloid production. Transformedhairy rootswere grown inGamborg’sB5
solid medium (half-strength basal salts, full-strength vitamins, 30g l21 sucrose,
6 g l21 agar, pH 5.7). Thetotal chloride concentration inGamborg’s B5formulation
was ,1 mM. We ground three-week-old hairy roots with a mortar, pestle and 106-
mm acid-washed glass beads in methanol (10 ml g21 fresh weight of hairy roots).
The crude natural product mixtures were filtered through 0.2-mm cellulose acetate
membrane (VWR) and subsequently subjected to LC–MS analysis. Hairy roots
transformed with wild-type A. rhizogenes lacking the plasmid were also evaluated.
Brominatedalkaloid production in transgenic C. roseus hairy roots. Wegrew a
selected transformed hairy root line for two weeks in low-chloride solid medium
(67mg l21 (NH4)2SO4, 353mg l21 Ca(NO3)2?4H2O, 61 mg l21 MgSO4, 1,250 mg
l21 KNO3, half-strength Murashige and Skoog micronutrient salts and full-
strength Murashige and Skoog vitamins, 3 mM total chloride concentration).
Hairy roots were transferred to the same medium supplemented with either pot-
assium bromide or potassium iodide (10–20 mM final concentration) and culti-
vated for an additional two weeks. They were then processed and alkaloid
production was analysed as described above (Supplementary Figs 12–15). We
performed experiments in duplicate.
22. Loris, E. A. et al. Structure-based engineering of strictosidine synthase: auxiliary for
alkaloid libraries. Chem. Biol. 14, 979–985 (2007).
23. Menzies, J. R., Paterson, S. J., Duwiejua, M. & Corbett, A. D. Opioid activity of
alkaloids extracted from Picralima nitida (fam. Apocynaceae). Eur. J. Pharmacol.
350, 101–108 (1998).
´
´
24. Frederich, M. et al. Antiplasmodial activity of alkaloids from various strychnos
species. J. Nat. Prod. 65, 1381–1386 (2002).
25. Zhu, W.-M. et al. Components of stem barks of Winchia calophylla A. DC. and their
bronchodilator activities. J. Integr. Plant Biol. 47, 892–896 (2005).
26. Amat, M., Linares, A. & Bosch, J. A new synthetic entry to pentacyclic Strychnos
alkaloids. Total synthesis of (.12.)-tubifolidine, (.12.)-tubifoline, and
(.12.)-19,20-dihydroakuammicine. J. Org. Chem. 55, 6299–6312 (1990).
´
27. Gandar, J. C. & Nitsch, C. Isolement de l’ester methylique d’un acide chloro-3-
´
`
indolylacetique a partir de graines immatures de pois, Pisum sativum L. C. R. Acad.
Sci. (Paris) Ser. D 265, 1795–1798 (1967).
28. Marumo, S., Hattori, H., Abe, H. & Munakata, K. Isolation of 4-chloroindolyl-3-acetic
acid from immature seeds of Pisum sativum. Nature 219, 959–960 (1968).
29. Deb Roy, A., Gru¨schow, S., Cairns, N. & Goss, R. J. Gene expression enabling
synthetic diversification of natural products: chemogenetic generation of
pacidamycin analogs. J. Am. Chem. Soc. 132, 12243–12245 (2010).
30. Li, S. et al. Assessment of the therapeutic activity of a combination of almitrine and
raubasine on functional rehabilitation following ischaemic stroke. Curr. Med. Res.
Opin. 20, 409–415 (2004).
Supplementary Information is linked to the online version of the paper at
www.nature.com/nature.
Acknowledgements We acknowledge support from the NIH (GM074820) and the
American Cancer Society (RSG-07-025-01-CDD). We thank H.-Y. Lee and M. Tjandra
for assistance with NMR characterizations and L. Li for high-resolution mass
spectroscopy analysis.
Full Methods and any associated references are available in the online version of
the paper at www.nature.com/nature.
Author Contributions All authors contributed to experimental design and data
analysis. X.Q. initiated the project and its design, and performed steady-state kinetics.
W.R. developed and implemented the transformation strategy and performed
steady-state kinetics and metabolite analysis. All authors contributed to the
preparation of the manuscript.
Received 26 April; accepted 20 September 2010.
Published online 3 November 2010.
1. Neumann, C. S., Fujimori, D. G. & Walsh, C. T. Halogenation strategies in natural
products biosynthesis. Chem. Biol. 15, 99–109 (2008).
2. Gribble, G. W. The diversity of naturally produced organohalogens. Chemosphere
52, 289–297 (2003).
3. Vaillancourt, F. H., Yeh, E., Vosburg, D. A., Garneau-Tsodiova, S. & Walsh, C. T.
Nature’s inventory of halogenation catalysts: oxidative strategies predominate.
Chem. Rev. 106, 3364–3378 (2006).
Author Information Reprints and permissions information is available at
www.nature.com/reprints. The authors declare no competing financial interests.
Readers are welcome to comment on the online version of this article at
www.nature.com/nature. Correspondence and requests for materials should be
addressed to S.E.O’C. (soc@mit.edu).
4 6 4
| N A T U R E | V O L 4 6 8 | 1 8 N O V E M B E R 2 0 1 0
Macmillan Publishers Limited. All rights reserved
©2010

LETTER RESEARCH
aerocolonigenes ATCC 39243 with forward and reverse PCR primers CrRebH that
introduce sites for SpeI and NcoI at the 59 end, and SpeI and PmlI at the 39 end
(underlined): 59-AAGACTACTAGTCCATGGATTCCGGCAAGATTGAC-39
and 59-ACTACTAGTCACGTGTCAGCGGCCGTGCTGTTGCC-39.
CaMV35S:Gus:NosPolyA, PyrH, RebH and RebF PCR fragments were indi-
vidually ligated into pGEM-Teasy vector (Promega) to yield pGEMCaMV35S,
pGEMPyrH, pGEMRebH and pGEMRebF, respectively.
METHODS
Heterologous expression and purification of C. roseus tryptophan decarbox-
ylase. The tryptophan decarboxylase (TDC) gene (accession number M25151.1)
was obtained by reverse-transcription PCR (RT–PCR) amplification of mRNA
isolated from C. roseus hairy root culture (Invitrogen, Dynabeads mRNA direct
kit) with PCR primers that introduce sites for NdeI and XhoI (underlined): 59-
AAAAAACATATGGGCAGCATTGATTCAACA-39 and 59-AAAAAACTCGA
GTCAAGCTTCTTTGAGCAAATC-39. The PCR fragment was subcloned into
the pGEM-T Easy Vector (Promega), and then excised and ligated into the NdeI/
XhoI site of the pET28-a plasmid (Novagen). The resulting pET28a–TDC con-
struct was subsequently transformed into BL21 (DE3) pLysS electrocompetent
Escherichia coli (Promega). A single E. coli colony harbouring pET28a–TDC was
inoculated in 5 ml lysogeny broth medium supplemented with kanamycin
(0.05 mg l21) and incubated overnight at 37 uC with shaking at 225r.p.m. An
aliquot of the overnight culture (1 ml) was then used to inoculate 100ml lysogeny
broth medium supplemented with kanamycin (0.05 mg l21) and incubated at
37 uC with shaking at 225r.p.m. until OD600 0.6 was reached. Cells were induced
for overexpression by the addition of isopropyl-b-D-galactopyranoside (IPTG;
final concentration, 1 mM) and the culture was allowed to continue growth for
16 h at 18 uC. Cells were harvested by centrifugation and lysed by sonication. The
hexahistidine-tagged TDC was purified using Ni-NTA Spin Kit (Qiagen) using
manufacturer’s protocols (Supplementary Fig. 1). Eluted enzyme was subse-
quently buffer-exchanged into phosphate buffer (50 mM NaH2PO4, 100mM
NaCl, pH 8.0) and immediately assayed for activity. This enzyme was not stable
after extended storage.
(v) Codon-optimization for C. roseus was performed to ensure efficient expression
of the prokaryotic halogenase and flavin reductase genes in plant cell culture.
Using codon usage database software (http://www.kazusa.or.jp/codon/), the fol-
lowing codons were identified as occurring at low frequency (triplet, frequency per
thousand): GCG, 4.8; CGG, 3.3; ACG, 4.0; TCG, 6.1; CCG, 6.0; and CCC, 6.7. Site-
directed mutagenesis was performed using a Stratagene QuikChange Site-
Directed Mutagenesis kit to replace rare codons with the most frequently occur-
ring codons encoding the corresponding amino acids. Only codons that appeared
within the first 300 nucleotides of the genes were subjected to mutagenesis.
The site-directed-mutagenesis primers (name, sequence) were as follows (the
sites of mutation are underlined): PyrH-SDM-for1, 59-GTGGGTGGTGGC
ACTGCTGGCTGGATGACC-39; PyrH-SDM-rev1, 59-GGTCATCCAGCCAGC
AGTGCCACCACCCAC-39; PyrH-SDM-for2, 59-GACATGCGGCCGTACAC
TACTGCTACCGCGATGAGCGCCGGC-39; PyrH-SDM-rev2, 59-GCCGGCG
CTCATCGCGGTAGCAGTAGTGTACGGCCGCATGTC-39; RebH-SDM-for,
59-CCCCAATCTGCAGACTGCTTTCTTCGACTTCCTCGGA-39; RebH-SDM-
rev, 59-TCCGAGGAAGTCGAAGAAAGCAGTCTGCAGATTGGGG-39; RebF-
SDM-for1, 59-ACCGCGGCCGATCACAGGGCTCTGATGAGCCTGTTTCCC-
39; RebF-SDM-rev1, 59-GGGAAACAGGCTCATCAGAGCCCTGTGATCGGCC
GCGGT-39; RebF-SDM-for2, 59-CTCGTCTGCCTGAACAGGGCTAGCGGAA
CGTTGCAC-39; RebF-SDM-rev2, 59-GTGCAACGTTCCGCTAGCCCTGTTC
AGGCAGACGAG-39.
Determining the steady-state kinetic constants of TDC for tryptophan sub-
strate analogues, 5- and 7-chlorotryptophan (1b and 1a). Steady-state kinetic
constants of TDC for 5- and 7-chlorotryptophan (1b and 1a) (Amatek) were
determined in phosphate buffer (0.1 M NaH2PO4, 3.5 mM b-mercaptoethanol,
pH 8.5) containing 1 mM pyridoxal-59-phosphate at 30 uC (0.3-ml reaction
volume) with TDC concentrations appropriate for obtaining the initial rate of
the reaction (0.6–0.9 mM). Aliquots (25ml) were quenched in 1 ml methanol,
containing yohimbine (500 nM) as an internal standard, at appropriate time
points. The samples were centrifuged (13,000 r.p.m. (16,000g), 5 min) to remove
particulates and then analysed by LC–MS. Samples were ionized by ESI with a
Micromass LCT Premier TOF Mass Spectrometer. The liquid chromatography
was performed on an Acquity Ultra Performance BEH C18, 1.7 mm, 2.13 100mm
column on a gradient of 10–90% acetonitrile/water (0.1% formic acid) over 5 min
at a flow rate of 0.6 ml min21. The appearance of the corresponding tryptamine
analogues (either 5- or 7-chlorotryptamine) was monitored by peak integration
and normalized to the internal standard. 5-chlorotryptamine was obtained from a
commercial source (Alfa Aesar). 7-chlorotryptamine was synthesized as prev-
iously reported3. Eight substrate concentrations (200–2,500 mM) were tested for
7-chlorotryptophan substrate, and six substrate concentrations (200–1,200 mM)
were tested for 5-chlorotryptophan substrate. Each concentration was assayed
three times and the average values are reported with standard deviations. The data
were fitted using nonlinear regression to the Michaelis–Menten equation using
ORIGINPRO 7 (OriginLab). For reference, the kinetic constants for the natural
substrate tryptophan (1) were also measured at concentrations ranging from 15 to
350mM (Supplementary Fig. 2).
Construction of halogenase plant expression vectors PyrH–RebF–STRvm and
RebH–RebF. The construction of plant expression vectors is summarized below
and in Supplementary Fig. 3.
(i) The CaMV35S:Gus:NosPolyA fragment was obtained by PCR amplification of
pCAMBIA1305.1 (Cambia) with forward and reverse PCR primers CaMV35S-
NosPolyA that introducesites for XbaI and KpnI at the 59 end, and PstIand SpeI at
the 39 end (underlined): 59-ACTTCTAGAGGTACCGGATCCTCTAGAGTCG
ACCTGCAG-39 and 59-ATTCTGCAGACTAGTCCCGATCTAGTAACATAG
ATGACACCG-39.
(ii) The tryptophan 5-halogenase gene (pyrH; accession number AAU95674) was
obtained by PCR amplification of genomic DNA isolated from Streptomyces
rugosporus NRRL 21084 with forward and reverse PCR primers CrPyrH that
introduce sites for XhoI and NcoI at the 59 end, and SpeI and BstEII at the 39
end (underlined): 59-ACTCTCGAGCCATGGATATCCGATCTGTGGTGATCG-
39 and 59-ACTACTAGTGGTAACCTCATTGGATGCTGGCGAGGTA-39.
(iii) The flavin reductase gene (rebF; accession number BAC15756) was obtained
by PCR amplification of genomic DNA isolated from Lechevalieria aerocoloni-
(vi) pGEMCaMV35S was digested and ligated into the KpnI/EcoRI sites of pSP72
vector (Promega) to yield pSPCaMV35S.
(vii) pGEMRebH and pGEMRebF were digested and ligated into the NcoI/PmlI
sites of pSPCaMV35S to yield pSPRebH and pSPRebF, respectively. Similarly,
pGEMPyrH was digested and ligated into the NcoI/BstEII sites of
pSPCaMV35S to yield pSPPyrH. pSPRebF was then digested and ligated into
the PstI site of pSP72 to yield pSPRebF_2.
(viii) pSPRebH and pSPPyrH were digested and ligated into the XbaI/EcoRI sites
of pSPRebF_2 to yield pSPRebHRebF and pSPPyrHRebF, respectively. Finally,
both pSPRebHRebF and pSPPyrHRebF were digested and ligated into the SpeI site
of pCAMBIA1300A to yield pCAMRebHRebF and pCAMPyrHRebF.
pCAMBIA1300A was constructed by introducing an SpeI restriction into
pCAMBIA1300 (Cambia). The site-directed-mutagenesis primers are (SpeI site
underlined) 59-CCCGCCTTCAGTTTAAACTAGTCAGTGTTTGACAGGAT-
39 and 59-atcctgtcaaacactgACTAGTttaaactgaaggcggg-39.
pCAMSTRvm was constructed as previously described21.
Generation of transgenic C. roseus hairy root cultures. The plant expression
construct pCAMRebHRebF was transformed into A. rhizogenes ATCC 15834 by
means of electroporation (1-mm cuvette, 1.25 kV). pCAMPyrHRebF and
pCAMSTRvm were co-transformed into A. rhizogenes ATCC 15834 via electro-
poration (1mm cuvette, 1.25 kV). Transformation of C. roseus seedlings with the
generated Agrobacterium strains was performed as previously reported31. Briefly,
180–250 C. roseus seedlings (Vinca Little Bright Eyes, Nature Hills Nursery) were
germinated aseptically on Gamborg’s B5 medium (full-strength basal salts, full-
strength vitamins, 30 g l21 sucrose, pH 5.7) and grown in a 16-h light, 8-h dark
cycle at 26 uC for 3 weeks. Seedlings were then wounded with extra-fine forceps at
the stem tip, and 3–5 ml A. rhizogenes from a freshly grown liquid culture were
inoculated on the wound.
Hairy roots appeared at the wound site 2–3 weeks after infection for about 80%
of the seedlings infected. After hairy roots reached 1–4 cm in length (usually about
6 weeks after infection), they were excised and transferred to Gamborg’s B5 solid
medium (half-strength basal salts, full-strength vitamins, 30 g l21 sucrose, 6 g l21
agar, pH 5.7) containing hygromycin (0.03 mg ml21) for selection and the anti-
biotic cefotaxime (0.25 mg ml21) to remove remaining bacteria. The total chloride
concentration in Gamborg’s B5 formulation was 1 mM. Allcultures were grown in
the dark at 26 uC. After the hygromycin selection process, hairy roots were main-
tained in solid medium lacking both hygromycin and cefotaxime.
To adapt the line to liquid culture, approximately 200mg of hairy roots (typ-
genes ATCC 39243 with forward and reverse PCR primers CrRebF that introduce ically five 3–4-cm-long stem tips) from each line that grew successfully on solid
sites for XhoI and NcoI at the 59 end, and SpeI and PmlI at the 39 end (underlined): medium were transferred to 50 ml of half-strength Gamborg’s B5 liquid medium.
59-ACTCTCGAGCCATGGATACGATCGAGTTCGACAGAC-39 and 59-ACT The cultures were grown at 26 uC in the dark at 125r.p.m. Hairy root growth in
ACTAGTCACGTGTCATCCCTCCGGTGTCCACAC-39.
liquid medium seemed to be slower than that in solid medium. Hairy root trans-
(iv) The tryptophan 7-halogenase gene (rebH; accession number BAC15758) was formants were screened for survival in solid medium supplemented with hygro-
obtained by PCR amplification of genomic DNA isolated from Lechevalieria mycin. The number of transformants decreased significantly after solid medium
Macmillan Publishers Limited. All rights reserved
©2010

RESEARCH LETTER
selection for each of the constructs transformed. Every line that grew in the of 10–90%acetonitrile/water (0.1% TFA) over 5 min at a flow rate of 0.6 mlmin21
.
selection medium was analysed for alkaloid production.
The capillary and sample cone voltages were 1,300 and 60 V, respectively. The
Hairy root selection and adaptation processes. For the plasmid pCAMRebH/ desolvation and source temperature were 300 and 100uC, respectively. The cone
RebF, the number of transformed hairy roots was 200 and the number of hairy and desolvation gas flow rates were 60 and 800l per hour, respectively. Analysis
roots after solid medium selection was 31. For the plasmid pCAMPyrH/RebF/
STRvm, the number of transformed hairy roots was 140 and the number of hairy
roots after solid medium selection was 57.
Verification of transferred DNA integration by genomic DNA analysis. To
verify the integration of transferred DNA (T-DNA) into the plant genome, the
genomicDNAfrom transformed hairyroots wasisolated (QiagenDNeasykit) and
then subjected to PCR amplification using T-DNA-specific primers with TDC
wasperformed with MASSLYNX4.1. Accurate mass measurements were obtained
in W-mode. The spectra were processed using the MASSLYNX 4.1 mass measure,
in which the mass spectrum of peaks of interest was smoothed and centred with
TOF mass correction, locking on the reference infusion of reserpine. Data for
RebF–RebH lines are shown in extracted LC–MS chromatograms in
Supplementary Figs 6–10.
Feeding of 7-chlorotryptamine (2a) in control C. roseus hairy root cultures
transformed with no plasmid. Alkaloid accumulation levels in hairy roots trans-
formed with RebH and RebF were compared with alkaloid accumulation levels in
control hairy root fed with 7-chlorotryptamine. The control hairy root line was
grown for 2 weeks in half-strength Gamborg’s B5 solid medium. Hairy roots were
then transferred to the same medium supplemented with 7-chlorotryptamine (2a;
0, 25, 50, 100, 200 and 750 mM final concentrations) and grown for a further 1
week. Hairy roots were then processed and alkaloid production analysed as
described in the previous subsection (Supplementary Fig. 16). Feeding studies
were performed in duplicate.
Brominated alkaloid production in transgenic C. roseus hairy roots. A selected
transformed hairy root line was grown for 2 weeks in low-chloride solid medium
(67mg l21 (NH4)2SO4, 353mg l21 Ca(NO3)2?4H2O, 61 mg l21 MgSO4, 1,250 mg
l21 KNO3, half-strength Murashige and Skoog micronutrient salts and full-
strength Murashige and Skoog vitamins, 3 mM total chloride concentration).
Hairy roots were transferred to the same medium supplemented with either pot-
assium bromide or potassium iodide (10–20 mM final concentration) and culti-
vated for an additional 2 weeks. Hairy roots were then processed and alkaloid
production analysed as described in the previous subsection but one. Hairy roots
transformed with wild-type A. rhizogenes lacking the plasmid were also evaluated
(Supplementary Figs 19–22). Experiments were performed in duplicate.
Purification and isolation of alkaloids from transformed TDC suppressed
hairy roots supplemented with 7-chlorotryptamine and 7-bromotryptamine.
To obtain chlorinated and brominated alkaloid standards, root tips (10–15) from
TDC suppressed hairy roots32 were subcultured in six 50 ml Gamborg’s B5 liquid
medium (half-strength basal salts, full-strength vitamins, 30 g l21 sucrose, pH 5.7)
and grown at 26 uC in the dark at 125r.p.m. for 3 weeks before the medium was
supplemented with either 7-chlorotryptamine (2a) or 7-bromotryptamine (2c)
(750 mM final concentration). Both tryptamine analogue substrates were synthe-
sized as previously reported33. After 2 weeks of co-cultivation, hairy roots were
extracted as described above in methanol (10ml g21 fresh hairy root weight).
Alkaloid extracts were filtered, concentrated under vacuum and redissolved in
25% acetonitrile/water (0.1% TFA) (1 ml g21 of fresh hairy root weight).
For cultures supplemented with 7-chlorotryptamine (2a), the redissolved mix-
ture was purified on a 103 20 mm Vydec reverse-phase column using a gradient
of 25–52% acetonitrile/water (0.1% TFA) over 24 min. Alkaloids were monitored
at 228nm and fractions containing the alkaloid analogues of interest, as deter-
mined by the characteristic isotopic distribution expected for chlorinated mole-
cules (35Cl/37Cl) from LC–MS analysis, were combined and concentrated under
vacuum (Supplementary Fig. 27).
For cultures supplemented with 7-bromotryptamine (2c), similar procedures
were performed to isolate alkaloids from transgenic hairy roots, except that the
liquid chromatography method was extended to 26 min. Alkaloids were moni-
tored at 228nm and fractions containing the alkaloid analogues of interest, as
determined by LC–MS analysis, were combined and concentrated under vacuum
(Supplementary Fig. 28).
Isolated alkaloids from both feedings were analysed by LC–MS (same para-
meters as above), analytical high-performance liquid chromatography and, where
possible, high resolution LC–MS (Supplementary Table 1), ultraviolet–visible
spectroscopy (Supplementary Fig. 29), tandem MS–MS (Supplementary Fig. 30)
and 1H NMR, 13C NMR and 1H–13C HSQC using a Bruker AVANCE-600 NMR
spectrometer equipped with a 5-mm 1H{13C,31P} cryoprobe (Supplementary
Figs 31 and 32). Halogenated alkaloids generally displayed longer retention times
than the natural alkaloids.
Quantification of chlorinated alkaloid production in transformed hairy roots.
12-chloro-19,20-dihydroakuammicine (5a) and 12-bromo-19,20-dihydroakuam-
micine (5c) standard curves were constructed by quantifying the peak areas of
several concentrations (20–1,400 nM) of each alkaloid authentic standard using
MASSLYNX 4.1. Similarly, 10-chloroajmalicine (6b), 15-chlorotabersonine (7b)
and 10-chlorocatharanthine (8b) standard curves were constructed by quantifying
the peak areas of several concentrations (20–1,400 nM) of each natural (that is,
primers serving as
a positive control (see below). Specifically, for the
pCAMRebHRebF hairy roots, primers for PCR amplification were designed to
amplify the complete TDC gene (TDC_for and TDC_rev), a 660-base-pair (bp)
region of the RebH gene (RebH_for and RebH_rev), a 680-bp region of the RebF
gene (RebF_for and RebF_rev) and an 800-bp region of the selection marker HPT
gene (HPT_for and HPT_rev) (see below).
PCR primers for verification of T-DNA integration of transformed hairy roots
were as follows (name, sequence): TDC_for, 59-AAAAAACATATGGGCAGC
ATTGATTCAACA-39; TDC_rev, 59-AAAAAACTCGAGTCAAGCTTCTTTG
AGCAAATC-39; RebH_for, 59-GTCTTCGATGCCGACCTCTTC-39; RebH_rev,
59-GTACATGTCGATCTTCTCCTGC-39; RebF_for, 59-TAGAGGACCTAACAG
AAC-39; RebF_rev, 59-CGTGACACTGGTCAGGGA-39; HPT_for, 59-GCCTGA
ACTCACCGCGACGTC-39; HPT_rev, 59-CCTCCAGAAGAAGATGTTGGC-39.
PCR amplification of genomic DNA from all of the selected transformed lines
(pCAMRebH/RebF cell line 4, lanes 1–4; pCAMRebH/RebF cell line 5, lanes 5–8;
pCAMRebH/RebF cell line 6, lanes 9–12; pCAMRebH/RebF cell line 10, lanes 13–
16; and pCAMRebH/RebF cell line 11, lanes 17–20) was successful for all four sets
of primers (Supplementary Fig. 4). PCR amplification of hairy root transformed
with A. rhizogenes lacking the pCAMBIA vector (provided by Professor Jacqueline
Shanks (Iowa State University) and Professor Carolyn Lee-Parsons (Northeastern
University)) genomic DNA was successful only when TDC-specific primers were
used (lanes 21–24). These results indicated that rebH and rebF were successfully
incorporated into the C. roseus genome in all chosen lines.
For the pCAMPyrH/RebF/STRvm hairy roots, primers for PCR amplification
were designed to amplify the complete TDC gene (TDC_for and TDC_rev), the
complete PyrH gene (PyrH_for and PyrH_rev), a 680-bp region of the RebF gene
(RebF_for and RebF_rev), a 440-bp region of the STRvm gene and an 800-bp
region of the selection marker HPT gene (HPT_for and HPT_rev) (see below and
Supplementary Fig. 5).
PCR primers for verification of T-DNA integration of transformed hairy roots
were as follows (name, sequence): TDC_for, 59-AAAAAACATATGGGC
AGCATTGATTCAACA-39; TDC_rev, 59-AAAAAACTCGAGTCAAGCTTCT
TTGAGCAAATC-3; PyrH_for, 59-ATGATCCGATCTGTGGTG-3; PyrH_rev,
59-TCATTGGATGCTGGCGAG-3; RebF_for, 59-TAGAGGACCTAACAGAAC-
3; RebF_rev, 59-CGTGACACTGGTCAGGGA-3; STRvm_for, 59-CCTTATTATT
GAAAGAGCTACATATG-3; STRvm_rev, 59-GCTAGAAACATAAGAATTTCC
CTTG-3; HPT_for, 59-GCCTGAACTCACCGCGACGTC-3; HPT_rev, 59-CCTCC
AGAAGAAGATGTTGGC-3.
PCR amplification of genomic DNA from three of four of the selected trans-
formed lines (pCAMPyrH/RebF/STRvm cell line 1, lanes 1–5; pCAMPyrH/RebF/
STRvm cell line 3, lanes 6–10, pCAMPyrH/RebF/STRvm cell line 6, lanes 11–15)
was successful for all five sets of primers (Supplementary Fig. 5). PCR amplifica-
tion of genomic DNA from pCAMPyrH/RebF/STRvm cell line 7 (lanes 16–20)
was successful when TDC, PyrH, RebF and HPT primers were used but not when
STRvm primers were used. PCR amplification of hairy root transformed with A.
rhizogenes lacking the pCAMBIA vector genomic DNA was successful only when
TDC specific primers were used (lanes 21–25).
Evaluation of alkaloid production in transgenic C. roseus hairy roots. Every
transgenic hairy root line that survived hygromycin selection medium was eval-
uated for alkaloid production. Transformed hairy roots were grown in Gamborg’s
B5 solid medium (half-strength basal salts, full-strength vitamins, 30 g l21 sucrose,
6 g l21 agar, pH 5.7). The total chloride concentration in Gamborg’s B5 formula-
tion was ,1 mM. Three-week-old hairy roots were ground with a mortar, pestle
and 106mm acid-washed glass beads in methanol (10 ml g21 of fresh hairy root
weight). The crude natural product mixtures were filtered through 0.2-mm cel-
lulose acetate membrane (VWR) and subsequently subjected to LC–MS analysis.
Additionally, hairy roots transformed with wild-type A. rhizogenes lacking the
plasmid were also evaluated.
These crude alkaloid mixtures were diluted 30:830 with methanol for mass
spectral analysis. Samples were ionized by ESI with a Micromass LCT Premier
TOF Mass Spectrometer. The liquid chromatography was performed on an
Acquity UltraPerformance BEH C18, 1.7 mm, 2.13 100mm columnona gradient non-chlorinated) alkaloid authentic standard using MASSLYNX 4.1.
Macmillan Publishers Limited. All rights reserved
©2010

LETTER RESEARCH
Dependence of chlorinated alkaloid production on concentrations of sodium
chloride. A selected transformed hairy root line was grown for 2 weeks in low-
chloride solid medium (67 mg l21 (NH4)2SO4, 353mg l21 Ca(NO3)2?4H2O,
61 mg L21 MgSO4, 1,250 mg l21 KNO3, half-strength Murashige and Skoog
micronutrient salts and full-strength Murashige and Skoog vitamins, 3 mM total
chloride concentration). Hairy roots were transferred to the same medium sup-
plemented with potassium chloride (0–20 mM final concentration) and grown for
a further 2 weeks. Hairy roots were then processed and alkaloid production was
analysed as previously described (Supplementary Figs 17 and 18).
GCATCTACTTCGTCT-39; RebH_rev, 59-TCGAACATCGTCTCGATCTC-39
(amplicon size, 117); RebF_for, 59-CTGATGAGCCTGTTTCCCA-39; RebF_rev,
59-CGTGACACTGGTCAGGGA-39 (amplicon size, 99); Rbps9_for, 59-TTGAGC
CGTATCAGAAATGC-39; Rbps9_rev, 59-CCCTCATCAAGCAGACCATA-39
(amplicon size, 122).
Real time RT–PCR was used to assess the expression levels of PyrH, RebF and
STRvm. Expression levels in hairy roots infected with A. rhizogenes lacking the
pCAMPyrH/RebF/STRvm construct were compared with expression levels in
hairy roots harbouring pCAMPyrH/RebF/STRvm (Supplementary Fig. 24).
Chlorinated alkaloid production in a line lacking STRvm expression is shown in
Supplementary Fig. 15. Photographs of the transformed roots are shown in
Supplementary Fig. 25.
PCR primers for real-time RT–PCR of pCAMPyrH/RebF/STRvm transformed
hairy roots were designed using GenScript web tool, and were as follows (name,
sequence): PyrH_for, 59-GCCTGCTCATCAACCAGAC-39; PyrH_rev, 59-CATC
GCGGTAGCAGTAGTGT-39 (amplicon size, 137); RebF_for, 59-CTGATGAG
CCTGTTTCCCA-39; RebF_rev, 59-CGTGACACTGGTCAGGGA-39 (amplicon
size, 99); STRvm_for, 59-TATTATTGAAAGAGCTACATATG-39; STRvm_rev,
59-CTCTGCACTGCCTTTCTTG-39 (amplicon size, 134); Rbps9_for, 59-TTGA
GCCGTATCAGAAATGC-39; Rbps9_rev, 59-CCCTCATCAAGCAGACCATA-
39 (amplicon size, 122).
Quantification of natural alkaloids in wild-type roots. The levels of natural
alkaloids in wild type hairy roots was quantified as described in section on quan-
tification of chlorinated alkaloid production in transformed hairy roots.
The levels of the four most abundant alkaloids found in these hairy roots,
ajmalicine (6), tabersonine (7), catharanthine (8), as well as tryptophan (1) and
tryptamine (2), were measured (Supplementary Fig. 26).
Assessment of the stability of chlorinated alkaloid production in subsequent
subcultures. Ten root tips from hairy roots transformed with pCAMRebH/RebF
and pCAMPyrH/RebF/STRvm were subcultured every 3 weeks in Gamborg’s B5
solid medium (half-strength basal salts, full-strength vitamins, 30 g l21 sucrose,
6g l21 agar, pH 5.7). and grown at 26 uC in the dark. Alkaloids were isolated from
21-day-old hairy roots and analysed as describedabove (Supplementary Figs 11–14).
Verification of expression of RebH, RebF and PyrH, STRvm enzymes by real-
time RT–PCR. Real-time RT–PCR was used to assess the expression levels of
RebH and RebF. Expression levels in hairy roots infected with A. rhizogenes
lacking the pCAMRebH/RebF construct were compared with expression levels
in hairy roots harbouring pCAMRebH/RebF. Messenger RNA from transformed
hairy roots was isolated and purified from contaminant DNA using a Qiagen
RNeasy Plant Mini Kit and Rnase-free DnaseI, respectively. The resulting
mRNA was then reverse-transcribed to cDNA using a Qiagen QuantiTect
Reverse transcription kit and then subjected to PCR with specific primers (see
below), a Qiagen SYBR Green PCR kit and a Biorad DNA Engine Opticon 2
system. The threshold cycle (CT) was determined as the cycle with a signal higher
than that of the background plus 10 s.d. Catharanthus roseus 40S ribosomal pro-
tein S9 (Rps9), encoded by a house-keeping gene, was used to adjust the amount of
the total mRNA in all samples. Real-time RT–PCR was performed in triplicate and
the data are pictured as the relative expression levels of rebH and rebF mRNA in
transgenic hairy roots as well as hairy roots lacking the pCAMBIA plasmid
(Supplementary Fig. 23).
31. Hughes, E. H. et al. Characterization of an inducible promoter system in
Catharanthus roseus hairy roots. Biotechnol. Prog. 18, 1183–1186 (2002).
32. Runguphan, W. et al. Silencing of tryptamine biosynthesis for production of
nonnatural alkaloids in plant culture Proc. Natl Acad. Sci. USA 106, 13673–13678
(2009).
33. Schumacher, R. W. et al. Synthesis of didemnolines A-D, N9-substituted
b-carboline alkaloids from the marine ascidian Didemnum sp. Tetrahedron 55,
935–942 (1999).
PCR primers for real-time RT–PCR of pCAMRebH/RebF transformed hairy
roots were designed using the GenScript web tool (http://www.genscript.com/ssl-
bin/app/primer), and were as follows (name, sequence): RebH_for, 59-GACGG
Macmillan Publishers Limited. All rights reserved
©2010