An enzymatic, stereoselective synthesis of (S)-norcoclaurine

Article abstract of DOI:10.1039/c0gc00036a

An efficient, stereoselective, green synthesis of (S)-norcoclaurine (higenamine) has been developed using the recombinant (S)-norcoclaurine synthase (NCS) enzyme, starting from the cheap tyrosine and dopamine substrates in a one-pot, two step process. Key steps in the biotransformation consist of the oxidative decarboxylation of tyrosine by stoichiometric amounts of sodium hypochlorite in order to generate 4-hydroxyphenylacetadehyde, followed by the addition of enzyme and dopamine substrate in the presence of ascorbate, a necessary ingredient in order to avoid oxidation of the catechol moiety. Quantitative extraction of the product from an aqueous solution was achieved by adsorption onto active charcoal dispersed in the reaction mixture. The optimized process afforded enantiomerically pure (S)-norcoclaurine (93%) in a yield higher than 80% and allowed good recovery of the enzyme for recycling. The process thus developed represents the first example of a green Pictet-Spengler synthesis, which may pave the way to novel strategies in benzylisoquinoline alkaloid synthesis.

Full text of DOI:10.1039/c0gc00036a

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/greenchem | Green Chemistry
An enzymatic, stereoselective synthesis of (S)-norcoclaurine†
Alessandra Bonamore,a,b Irene Rovardi,c Francesco Gasparrini,a,c Paola Baiocco,b Marco Barba,c
Carmela Molinaro,c Bruno Botta,a,c Alberto Boffia,b and Alberto Macone*b
Received 23rd April 2010, Accepted 29th June 2010
DOI: 10.1039/c0gc00036a
An efficient, stereoselective, green synthesis of (S)-norcoclaurine (higenamine) has been developed
using the recombinant (S)-norcoclaurine synthase (NCS) enzyme, starting from the cheap tyrosine
and dopamine substrates in a one-pot, two step process. Key steps in the biotransformation
consist of the oxidative decarboxylation of tyrosine by stoichiometric amounts of sodium
hypochlorite in order to generate 4-hydroxyphenylacetadehyde, followed by the addition of
enzyme and dopamine substrate in the presence of ascorbate, a necessary ingredient in order to
avoid oxidation of the catechol moiety. Quantitative extraction of the product from an aqueous
solution was achieved by adsorption onto active charcoal dispersed in the reaction mixture. The
optimized process afforded enantiomerically pure (S)-norcoclaurine (93%) in a yield higher than
80% and allowed good recovery of the enzyme for recycling. The process thus developed
represents the first example of a green Pictet–Spengler synthesis, which may pave the way to novel
strategies in benzylisoquinoline alkaloid synthesis.
between the aldehyde carbonyl and the phenylethyl amine
Introduction
moiety, followed by a Mannich-type cyclization to yield tetrahy-
Benzylisoquinoline alkaloids are among the most important
drobenzylisoquinoline racemic mixtures.5 In order to obtain
plant secondary metabolites since they include a number of
the biologically-active (S)-isomer, diverse synthetic strategies
biologically-active substances that are widely employed as phar-
have been employed based on asymmetric catalytic approaches
maceuticals. The rich bioactive potential of these compounds is
involving the enantioselective hydrogenation of the correspond-
being rigorously explored, as demonstrated by the recent start
ing dihydroisoquinoline intermediates by appropriate chiral
of clinical trials on higenamine ((S)-norcoclaurine) derivatives
for the treatment of septic shock syndrome.1 (S)-Norcoclaurine
is also an efficient b1-adrenergic drug, with a pharmacokinetic
profile that competes favourably with dobutamine, the currently
accepted best in its class. Nevertheless, the low availability of
natural (S)-norcoclaurine and the expensive stereoselective syn-
thesis of this compound with respect to dobutamine (currently
marketed as a racemic mixture) have hampered its clinical
development. (S)-Norcoclaurine is thus the first example of
a series of particularly interesting secondary metabolites of
the benzylisoquinoline pathway that may provide a handful
of therapeutically-useful natural products, once their synthetic
strategy has been redesigned on the basis of cheaper and cleaner
synthetic routes.
metal catalysts.1,2 These synthetic routes are highly efficient and
guarantee a very good enantioselectivity. Nevertheless, large
scale preparations entail the extensive use of organic solvents
and metal catalysts, which involve the need for solvent and
metal recycling. Alternatively, an enzymatic synthesis may offer
the advantage of a clean and green synthetic process in the
absence of organic solvents and metal catalysts, provided that
the efficiency of the process is at least comparable to that
of the traditional route. The enzymatic pathway leading to
benzylisoquinoline derivatives has been shown to originate from
a common route, in which the first committed step consists of the
stereospecific Pictet–Spengler condensation of dopamine with 4-
hydroxyphenylacethaldehyde (4-HPAA) to yield the benzyliso-
quinoline central precursor (S)-norcoclaurine.6 The enzyme (S)-
norcoclaurine synthase (NCS) has been recently identified7,8 and
shown to be an efficient bifunctional reaction catalyst that is
able to bind and activate the p-hydroxybenzaldehyde substrate
and subsequently drive the Mannich-type condensation with the
incoming dopamine substrate in an exquisitely enantioselective
manner.9–11 However, the reaction conditions for NCS catalysis
reported thus far have limited synthetic advantage in that they
are limited to analytical scale (S)-norcoclaurine preparation. In
particular, both the generation of the aldehyde and the reaction
with the dopamine substrate suffer from insufficient yields and
are affected by parasitic reactions, such as aldehyde polymer-
ization and dopamine oxidation, that take place in aqueous
solutions, thus rendering the whole synthesis impractical. In this
paper, the overall NCS-catalyzed reaction has been revisited in
The synthesis of benzylisoquinoline alkaloids is based on
the Pictet–Spengler reaction, which entails the acid-catalyzed
electrophilic addition of an iminium ion to a substituted benzyl
species.2–4 The reaction mechanism consists of a two step process,
in which the iminium ion is generated first from a condensation
aMOLIROM s.r.l, via Carlo Bartolomeo Piazza no 8, 00161, Rome, Italy
bDipartimento di Scienze Biochimiche and Istituto di Biologia e Patologia
Molecolari (IBPM), CNR, Universita` “Sapienza”, P. Aldo Moro 5,
00185, Rome, Italy. E-mail: alberto.macone@uniroma1.it
cDipartimento di Studi di Chimica e Tecnologia delle Sostanze
Biologicamente Attive, Universita` “Sapienza”, P. Aldo Moro 5, 00185,
Rome, Italy
† Electronic supplementary information (ESI) available: Further exper-
imental data. See DOI: 10.1039/c0gc00036a
This journal is
The Royal Society of Chemistry 2010
Green Chem., 2010, 12, 1623–1627 | 1623
©

View Article Online
order to provide an environmentally-friendly, easily scalable and
robust process that expands the scope of the biotransformation
to a multigram level. The present approach is a first attempt
to exploit the potential of a Pictet–Spengler enzyme in the
direct synthesis of chiral benzylisoquinolines, thus suggesting
that many other enzymes belonging to this family might be used
for the synthesis of precious plant secondary metabolites.
dehydrogenase starting from tyrosol were unsuccessful (data not
shown). In turn, the chemical synthesis of the pure aldehyde
is impractical in that it entails anhydrous preparations to be
stored at low temperature.12 Thus, 4-HPAA was conveniently
generated by the oxidative decarboxylation of tyrosine in
aqueous solutions in the presence of an equimolar amount of
hypochlorite, followed by its immediate use in the enzymatic
reaction. The product yield was estimated to be greater than
99%, with less than 0.1% chlorinated by-products (by GC-MS).
The reaction was carried out in the same phosphate buffer as
needed for NCS catalysis, thus suggesting that the whole (S)-
norcoclaurine synthesis could be optimized as a one pot reaction
by quickly adding dopamine and NCS to the solution containing
the newly synthesized aldehyde.
It is important to focus on the critical step concerning the
low stability of dopamine in air-equilibrated aqueous solutions,
which is responsible of the low overall yields of the intermediates
in benzylisoquinoline alkaloid synthesis.13 Dopamine is in fact
easily oxidized to melanine-like pigments in a second order
reaction with oxygen. The dopamine oxidation reaction severely
impairs the possibility of reaching synthetically convenient
concentrations higher than 1–2 mM. Thus, several attempts
were made in order to improve the effective concentration
of the dopamine substrate in solution, including de-aeration
(N2 purging into the reaction batch) or by the use of solvent
mixtures (water/isopropanol) in the presence of reductants
(sodium dithionite, menadiol, vitamin E). Among the different
reductants used, ascorbate was observed to be the most efficient,
easiest to use and cheapest. Thus, in order to prevent dopamine
oxidation during the reaction course, the reaction mixture had
5 mM ascorbate added in the absence of cosolvents. This strategy
allowed ten-fold higher dopamine concentrations (10 mM) to be
used.
Results and discussion
The present data highlight a method for the efficient en-
zymatic synthesis of (S)-norcoclaurine, the key precursor of
benzylisoquinoline alkaloids. In order to develop a green and
easily scalable process, a one-pot, two-step synthesis was set up
starting from tyrosine and dopamine in the presence of NCS
(Scheme 1).
On the basis of the experimental findings outlined above, (S)-
norcoclaurine synthesis was achieved from 10 mM dopamine
and 4-HPAA substrates in a yield of 2.2 g isolated product (81%
overall yield) from 1 L of solution and 30 min incubation at
37 ◦C. Product formation was followed by GC-MS, as reported
in Fig. 1 (see also the ESI, Fig. S1†). The enzyme concentration
was adjusted to achieve relatively fast reaction kinetics in order
to limit the competing non-enantioselective chemical coupling
between dopamine and 4-HPAA, which would dominate the
NCS reaction for longer periods of incubation. The optimal
enzyme concentration was found to be 0.5 mM, corresponding
to 10 mg L-1 enzyme in the solution.
In Fig. 2, the chiral HPLC chromatograms of standard,
racemic norcoclaurine and the reaction mixture are reported,
and the enantiomeric excess of (S)-norcoclaurine was found to
be 93%. This result was confirmed by CD analysis, indicating
the presence of two peaks with opposite Cotton effects (due
to the presence of both enantiomers) in standard racemic
norcoclaurine (Fig. 3A), whereas the reaction mixture shows
a very large excess of the expected (S)-isomer (Fig. 3B). The
newly synthesized (S)-norcoclaurine did not racemize after three
months storage as a dry powder at room temperature.
Scheme
1 The stereospecific chemoenzymatic synthesis of (S)-
norcoclaurine from tyrosine and dopamine. Reagents and conditions:
(i) NaClO, phosphate buffer pH 7.0, 1 h, 37 ◦C; (ii) NCS 0.5 mM,
ascorbate 5 mM, phosphate buffer 0.1 M pH 7.0, 0.5 h, 25 ◦C.
NCS was recombinantly expressed in Eschericha coli in high
yield using a codon-optimized synthetic gene (Geneart AG),
as described previously.9,10 E. coli cells were fermented in a 2
L Sartorius fermentor by a feed batch procedure in minimal
medium at 25 ◦C. The yield of wet bacterial paste was about 50 g
L-1 fermented medium with a raw yield of 30 mg protein per gram
of bacterial paste. The his-tagged protein was easily purified
in batch by a standard procedure on a nickel nitriloacetate
resin. The present expression and purification protocol allows
one to obtain ten-fold larger amounts of purified protein with
respect to previously reported methods,10 which consisted of low
density growth in shake flasks suitable only for analytical scale
preparation. The large amount of enzyme obtained represents
the pre-requisite for scaling-up (S)-norcoclaurine production.
The major limiting step in the large scale production of
(S)-norcoclaurine is represented by the low stability of both
dopamine and the aldehydic substrate in aqueous solution
at neutral pH values. In fact, 4-HPAA is not commercially
available, and all efforts to produce it enzymatically using
plant monoamino oxidases starting from tyramine or alcohol
An extraction strategy that avoids the use of organic solvents
was developed based on the use of activated carbon (NORIT,
multipurpose activated charcoal). The adsorption/desorption
properties of activated carbon towards phenolic compounds
1624 | Green Chem., 2010, 12, 1623–1627
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
Fig. 3 (A) Chiral HPLC chromatogram of a standard racemic mixture
of norcoclaurine. Acquisition was performed with a CD detector at
280 nm. (B) HPLC chromatogram of the reaction mixture, showing the
presence of the expected (S)-norcolaurine enantiomer (peak 1).
tion was achieved by shaking carbon granules directly added
to the aqueous phase at room temperature (30 min shaking).
Control experiments indicated that the dopamine substrate is
not absorbed onto the activated carbon beads, whereas the
aldehyde substrate is strongly absorbed. The best desorption
conditions for (S)-norcoclaurine were obtained in ethanol at
Fig. 1 (A) A GC-MS chromatogram of the reaction mixture after
extraction and derivatization with TMS-chloride. (B) The EI mass
spectrum of TMS-derivatized (S)-norcolaurine. The inset shows the two
major ions derived from TMS-(S)-norcoclaurine fragmentation (m/z =
308 and 179).
◦
40 C in the presence of a slight molar excess of NaOH. This
procedure allowed a very good yield of (S)-norcoclaurine to
be obtained in the absence of unreacted aldehyde contaminant,
which remained adsorbed on the carbon matrix. This yield is
comparable to that achieved by extraction with a four volume
excess of diethyl ether (81% vs. 79%). The procedure based
on activated carbon has the notable advantage of recovering
about 70% of the active NCS enzyme from the reaction mixture
once the carbon has been removed by filtration. In recycling
experiments (Fig. 4), the same enzyme solution could be used
up to five times, with an overall product recovery of about
Fig. 2 (A) Chiral HPLC chromatogram of a standard racemic mixture
of norcoclaurine. Acquisition was performed with a UV detector at
254 nm. (B) HPLC chromatogram of the reaction mixture, showing the
presence of the expected (S)-norcolaurine enantiomer (peak 1).
Fig. 4 Reaction recycling: Five recycling cycles were performed by
monitoring the enzyme activity after each step (white bars) and the
product yield (black bars). The enzyme concentration in centricon 20 mL
tubes was obtained before each cycle.
have been widely characterized,14 and conditions for optimal
binding/desorption of (S)-norcoclaurine were investigated as a
function of temperature and solvent composition. The absorp-
This journal is
The Royal Society of Chemistry 2010
Green Chem., 2010, 12, 1623–1627 | 1625
©

View Article Online
670 mg of product (isolated) per mg of enzyme. However, further
optimization may be necessary to make this recycling cost-
effective. The current limitation of the process resides mainly
in the relatively low stability of the aldehyde substrate. In turn,
the aldehyde might be conveniently generated in situ by means
of an appropriate amino acid or amine oxidase. With this aim
in mind, several enzymes bearing appropriate specificities for
tyrosine or tyramine are presently being tested.
synthase (10 mg, corresponding to a final concentration of
0.5 mM) were added to the reaction mixture in the presence
of ascorbate (5 mM) and incubated for 30 min at 37 ◦C.
Activated carbon NORIT (Sigma-Aldrich 93067) was used as
the adsorbent for the purification of (S)-norcoclaurine. Briefly,
10 g of adsorbent were added to the aqueous reaction mixture (1
L). After 30 min shaking at room temperature, the mixture was
filtered. The adsorbent was recovered and washed twice with
50 mL of distilled water. It was then transferred into a conical
flask and treated with 100 mL of ethanolic sodium hydroxide
(0.005 N NaOH in 99% ethanol). Desorption was carried out
by shaking for 2 h at 40 ◦C. The (S)-norcoclaurine-enriched
organic fraction was neutralized with 0.005 N HCl, evaporated
to dryness under reduced pressure and the resulting solid
Conclusions
In conclusion, a new, easily scalable process for the synthesis
of (S)-norcoclaurine has been reported according to a one-pot,
two-step reaction by using norcoclaurine synthase enzyme, pro-
duced in E. coli and purified in high yield. The stereoselectivity
of the reaction was 93% and the reaction yields were found to
be about 80%.
The results reported in this study provide an example of a
chemoenzymatic synthetic strategy as a novel, efficient and green
tool for manufacturing plant-derived metabolites, particularly
benzylisoquinoline alkaloids.
characterized by means of GC-MS, ESI/MS, 1H NMR and 13
C
NMR. Enantiomeric excesses were determined by chiral HPLC.
The specific rotations of MeOH solutions of norcoclaurine were
measured using a JASCO P-1030 polarimeter at 25 ◦C (cell path
10 cm).
ESI-MS
1 mg of (S)-norcoclaurine, obtained as described above, was
dissolved in 1 mL of acetonitrile and analyzed by ESI-MS
(Thermo Finnigan LXQ). The spectrometer had an electrospray
ion source and a linear ion trap analyzer. The ESI capillary
Experimental section
Chemicals
◦
temperature was 275 C and the analysis flow was 5 mL min-1.
Reagents obtained from commercial suppliers were used without
further purification. (R,S)-norcoclaurine was obtained from
Sequoia Research Products (UK).
The positive ESI-MS of norcoclaurine was observed with an
m/z = 272 [M + H]+.
NCS expression and purification
GC-MS analysis
A synthetic gene coding for norcoclaurine synthase (NCS)
from Thalictrum flavum has been constructed by GENEART
(GmbH, Germany) with optimised E. coli codons. The NCS
protein truncated at the first 19 amino acids with a His-tag
at the C-terminus was expressed in E. coli BL21 (DE3) cells
under fermentative conditions. The cells were harvested by
centrifugation (4000g for 10 min at 4 ◦C). Cell pellets were
frozen overnight, then resuspended and sonicated in 10 mL
of 50 mM phosphate buffer pH 8 containing 300 mM NaCl
(buffer A) and 1 mM phenylmethylsulfonyl fluoride (PMSF).
The extracts were centrifuged at 16000g for 20 min at 4 ◦C. NCS
was purified in batch using a nickel-chelating resin (Protino,
Ni-TED, Macherey-Nagel). After washing with buffer A, the
protein was eluted with buffer A containing 500 mM imidazole
(buffer B).
An Agilent 6850A gas chromatograph coupled to a 5973 N
quadrupole mass selective detector (Agilent Technologies, Palo
Alto, CA, USA) was used. Chromatographic separations were
carried out on an Agilent HP5 ms fused-silica capillary col-
umn (30 m ¥ 0.25 mm i.d.) coated with 5% phenyl 95%
dimethylpolysiloxane (film thickness 0.25 mm) as a stationary
phase. Injection mode: splitless at a temperature of 260 ◦C.
◦
Column temperature program: 100 ◦C (1 min), then to 300 C
at a rate of 15 ◦C min-1 and then held for 5 min. The carrier gas
was helium at a constant flow rate of 1.0 mL min-1. Spectra were
obtained in electron impact mode at 70 eV ionization energy
and a mass scan range from m/z = 50 to 500; ion source 280 ◦C;
ion source vacuum 10-5 Torr.
(S)-Norcoclaurine formation was followed by withdrawing
aliquots of the reaction mixture (20 mL) until the increase in
yield was negligible. Each aliquot was dried under an N2 stream
and directly derivatized with trimethylsilyl chloride in order to
obtain the corresponding trimethylsilyl ether.
One-pot preparation of norcoclaurine from
4-hydroxyphenylacetaldehyde and dopamine
4-Hydroxyphenylacetaldehyde was prepared according to the
method of Hazen et al.15 A solution of tyrosine (10 mmol) was
mixed with a solution of NaClO (10 mmol) in phosphate buffer
50 mM pH 7.0 (1 h at 37 ◦C) to a final volume of 1 L. Aliquots
of 50 mL of the reaction mixture were extracted with 200 mL of
diethyl ether and directly injected into the GC-MS instrument
(1 mL) to follow the aldehyde formation. Once tyrosine had been
completely converted in 4-hydroxyphenylacetaldehyde (10 mM
final concentration), dopamine (10 mmol) and norcoclaurine
HPLC analysis for enantiomeric excess determination
Separation of the norcoclaurine enantiomers was performed
by a HPLC analysis carried out on a chiral stationary phase
with Teicoplanin as the selector, obtained from Gasparrini
F. (CSP-Teicoplanin, 250 ¥ 4.0 mm); the mobile phase was
methanol–acetonitrile (70/30 v/v) containing 0.25% Et3N and
0.25% CH3COOH; flow rate: 1.00 mL min-1 at 25 ◦C. UV
and CD detections were performed at l = 254 and 280 nm,
1626 | Green Chem., 2010, 12, 1623–1627
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
respectively. The enantiomeric excess of the product was de-
termined by chiral HPLC. The retention times of the (S)- and
(R)-norcoclaurine isomers were 12.5 and 21.5 min, respectively.
The [a]2D5 value obtained was -24.7 in methanol. The compound
is thus assigned as (S)-(-) norcoclaurine, in agreement with
ref. 16.
References
1 M. K. Pyo, D. H. Lee, D. H. Kim, J. H. Lee, J. C. Moon, K. C. Chang
and H. S. Yun-Choi, Bioorg. Med. Chem. Lett., 2008, 18, 4110–4114.
2 J. Seayad, A. M. Seayad and B. List, J. Am. Chem. Soc., 2006, 128,
1086–1087.
3 L. Y. Luk, S. Bunn, D. K. Liscombe, P. J. Facchini and M. E. Tanner,
Biochemistry, 2007, 46, 10153–10161.
4 R. Noyori, K. Masato, T. Hidemasa, K. Hidenori, A. Sasumu, US
pat., 4851537, 1989.
5 B. E. Maryanoff, H. C. Zhang, J. H. Cohen, I. J. Turchi and C. A.
Maryanoff, Chem. Rev., 2004, 104, 1431–1628.
6 P. J. Facchini, Annu. Rev. Plant Physiol. Plant Mol. Biol., 2001, 52,
29–66.
NMR analysis
NMR measurements were carried out in CD3OD at room
temperature on a Varian Unity Inova instrument at 400 MHz
and 296 K. Chemical shifts are expressed in d values relative to
tetramethylsilane (TMS) as an internal reference.
1H NMR (CD3OD, 400 MHz): d 7.09 (d, J = 7.5 Hz; H-2¢,
H-6¢), 6.77 (d, J = 7.5 Hz; H-3¢, H-5¢), 6.64 (s, H-5), 6.55 (s, H-8),
4.21 (br s, H-1), 3.21 (m, H2-3), 2.89 (m, H2-9), 2.76 (m, H2-4);
13C NMR (CD3OD, 400 MHz): d 156.4 (C-4¢), 144.6, 143.8 (C-6,
C-7), 130.3 (C-2¢, C-6¢), 127.8, 126.1, 124.3 (C-4a, C-8a, C-1¢),
115.4 (C-3¢, C-5¢), 115.1, 113.0 (C-5, C-8), 56.7 (C-1), 40.3, 39.9
(C-3, C-9), 34.0 (C-4).
7 N. Samanani and P. J. Facchini, J. Biol. Chem., 2002, 277, 33878–
33883.
8 N. Samanani, D. K. Liscombe and P. J. Facchini, Plant J., 2004, 40,
302–313.
9 A. Pasquo, A. Bonamore, S. Franceschini, A. Macone, A. Boffi and
A. Ilari, Acta Crystallogr., Sect. F: Struct. Biol. Cryst. Commun.,
2008, 64, 281–283.
10 A. Ilari, S. Franceschini, A. Bonamore, F. Arenghi, B. Botta, A.
Macone, A. Pasquo, L. Bellucci and A. Boffi, J. Biol. Chem., 2008,
284, 897–904.
11 A. Bonamore, M. Barba, B. Botta, A. Boffi and A. Macone,
Molecules, 2010, 15, 2070–2078.
12 N. Samanani and P. J. Facchini, Planta, 2001, 213, 898–906.
13 H. Minami, J. S. Kim, N. Ikezawa, T. Takemura, T. Katayama, H.
Kumagai and F. Sato, Proc. Natl. Acad. Sci. U. S. A., 2008, 105,
7393–7398.
Acknowledgements
¨
14 B. Ozkaya, J. Hazard. Mater., 2006, 129, 158–163.
We gratefully acknowledge Prof. Rodolfo Federico for helpful
discussions and valuable suggestions. A Research Grant from
Fondazione Roma to B. B., and local university grants to A. B.
and B. B. are gratefully acknowledged.
15 S. L. Hazen, F. F. Hsu and J. W. Heinecke, J. Biol. Chem., 1996, 271,
1861–1867.
16 K. S. Chung, H. S. Yun-Choi and Y. Hahn, Nat. Prod. Sci., 2000, 6,
20–24.
This journal is
The Royal Society of Chemistry 2010
Green Chem., 2010, 12, 1623–1627 | 1627
©