Asymmetric synthesis of tetrahydroisoquinolines by enzymatic Pictet-Spengler reaction

Article abstract of DOI:10.1080/09168451.2014.890039

Norcoclaurine synthase (NCS) catalyzes the stereo-selective Pictet-Spengler reaction between dopamine and 4-hydroxyphenylacetaldehyde as the first step of benzylisoquinoline alkaloid synthesis in plants. Recent studies suggested that NCS shows relatively relaxed substrate specificity toward aldehydes, and thus, the enzyme can serve as a tool to synthesize unnatural, optically active tetrahydroisoquinolines. In this study, using an N-terminally truncated NCS from Coptis japonica expressed in Escherichia coli, we examined the aldehyde substrate specificity of the enzyme. Herein, we demonstrate the versatility of the enzyme by synthesizing 6,7-dihydroxy-1-phenethyl-1,2,3,4-tetrahydroisoquinoline and 6,7-dihydroxy-1-propyl-1,2,3,4-tetrahydroisoquinoline in molar yields of 86.0 and 99.6% and in enantiomer excess of 95.3 and 98.0%, respectively. The results revealed the enzyme is a promising catalyst that functions to stereoselectively produce various 1-substituted-1,2,3,4-tetrahydroisoquinolines.

Full text of DOI:10.1080/09168451.2014.890039

This article was downloaded by: [University of Newcastle (Australia)]
On: 13 September 2014, At: 07:46
Publisher: Taylor & Francis
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer
House, 37-41 Mortimer Street, London W1T 3JH, UK
Bioscience, Biotechnology, and Biochemistry
Publication details, including instructions for authors and subscription information:
http://www.tandfonline.com/loi/tbbb20
Asymmetric synthesis of tetrahydroisoquinolines by
enzymatic Pictet–Spengler reaction
Masakatsu Nishihachijoa, Yoshinori Hiraia, Shigeru Kawanoa, Akira Nishiyamaa, Hiromichi
Minamic, Takane Katayamac, Yoshihiko Yasoharab, Fumihiko Satod & Hidehiko Kumagaic
a QOL Division Fine Chemicals Group Research Team, Kaneka Corporation, Takasago,
Japan
b Frontier Biochemical and Medical Research Laboratories, Kaneka Corporation,
Takasago, Japan
c Research Institute for Bioresources and Biotechnology, Ishikawa Prefectural University,
Nonoichi, Japan
d Division of Integrated Life Science, Graduate School of Biostudies, Kyoto University,
Kyoto, Japan
Published online: 29 Apr 2014.
To cite this article: Masakatsu Nishihachijo, Yoshinori Hirai, Shigeru Kawano, Akira Nishiyama, Hiromichi Minami,
Takane Katayama, Yoshihiko Yasohara, Fumihiko Sato & Hidehiko Kumagai (2014) Asymmetric synthesis of
tetrahydroisoquinolines by enzymatic Pictet–Spengler reaction, Bioscience, Biotechnology, and Biochemistry, 78:4,
701-707, DOI: 10.1080/09168451.2014.890039
To link to this article: http://dx.doi.org/10.1080/09168451.2014.890039
PLEASE SCROLL DOWN FOR ARTICLE
Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained
in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no
representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of
the Content. Any opinions and views expressed in this publication are the opinions and views of the authors,
and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied
upon and should be independently verified with primary sources of information. Taylor and Francis shall
not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other
liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or
arising out of the use of the Content.
This article may be used for research, teaching, and private study purposes. Any substantial or systematic
reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any
form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://
www.tandfonline.com/page/terms-and-conditions

Bioscience, Biotechnology, and Biochemistry, 2014
Vol. 78, No. 4, 701–707
Asymmetric synthesis of tetrahydroisoquinolines by enzymatic Pictet–Spengler
reaction
Masakatsu Nishihachijo1,*, Yoshinori Hirai1, Shigeru Kawano1, Akira Nishiyama1,
Hiromichi Minami3, Takane Katayama3, Yoshihiko Yasohara2, Fumihiko Sato4 and
Hidehiko Kumagai3
2
1QOL Division Fine Chemicals Group Research Team, Kaneka Corporation, Takasago, Japan; Frontier
3
Biochemical and Medical Research Laboratories, Kaneka Corporation, Takasago, Japan; Research Institute for
4
Bioresources and Biotechnology, Ishikawa Prefectural University, Nonoichi, Japan; Division of Integrated Life
Science, Graduate School of Biostudies, Kyoto University, Kyoto, Japan
Received November 11, 2013; accepted December 2, 2013
http://dx.doi.org/10.1080/09168451.2014.890039
Norcoclaurine synthase (NCS) catalyzes the stereo-
selective Pictet–Spengler reaction between dopamine
and 4-hydroxyphenylacetaldehyde as the first step of
benzylisoquinoline alkaloid synthesis in plants.
Recent studies suggested that NCS shows relatively
relaxed substrate specificity toward aldehydes, and
thus, the enzyme can serve as a tool to synthesize
unnatural, optically active tetrahydroisoquinolines.
In this study, using an N-terminally truncated NCS
from Coptis japonica expressed in Escherichia coli,
we examined the aldehyde substrate specificity of the
enzyme. Herein, we demonstrate the versatility of
the enzyme by synthesizing 6,7-dihydroxy-1-phen-
ethyl-1,2,3,4-tetrahydroisoquinoline and 6,7-dihy-
in contrast to the chemical reaction, the NCS-catalyzed
enzymatic reaction proceeds stereoselectively
(Scheme 1). Therefore, if NCS can accommodate vari-
ous aldehyde molecules as the substrates, the enzyme
could serve as a promising biocatalyst for the produc-
tion of unnatural, optically active 1-substituted-1,2,3,4-
tetrahydroisoquinolines that are important starting mate-
rials in the pharmaceutical and agricultural fields.
Recently, synthesis of tetrahydroisoquinolines using
NCS from Thalictrum flavum (TfNCS) was reported.3,4)
TfNCS was shown to recognize phenylacetaldehydes
that are substituted with various electron-withdrawing
or donating groups, heteroaromatic moieties, aromatic
bicyclics, aliphatic cycles, and aliphatic open-chained
compounds. In that study, however, the stereoselectivity
of the reactions was not established, for example, enan-
tiomeric excess was not determined (note that the Pic-
tet–Spengler reaction occurs spontaneously in aqueous
solution).
We have previously cloned the NCS gene (CjPR10A;
GenBank Accession No. AB267399) from Coptis
japonica and used the enzyme to construct a bacterial
platform for fermentative production of (S)-reticu-
line.5−7) Herein, to further demonstrate the versatility of
the enzymatic Pictet–Spengler reaction, we examined
the aldehyde substrate specificity of CjNCS and estab-
lished an efficient method for synthesizing “optically
active” 1-substituted-1,2,3,4-tetrahydroisoquinolines.
droxy-1-propyl-1,2,3,4-tetrahydroisoquinoline
in
molar yields of 86.0 and 99.6% and in enantiomer
excess of 95.3 and 98.0%, respectively. The results
revealed the enzyme is a promising catalyst that
functions to stereoselectively produce various
1-substituted-1,2,3,4-tetrahydroisoquinolines.
Key words: norcoclaurine synthase; Pictet–Spengler
reaction; isoquinoline alkaloid
Isoquinoline alkaloids constitute a large group of
secondary metabolites of higher plants. They include
pharmaceutically important molecules such as the anal-
gesic compound morphine, the antibacterial agent ber-
berine, and the antispasmodic drug papaberine (Fig. 1).
Although these compounds show variable structures,
the synthetic pathway essentially involves (S)-norcocla-
urine as a common intermediate (Fig. 1).1) (S)-Norco-
claurine is produced by norcoclaurine synthase (NCS)
(EC 4.2.1.78) that catalyzes the condensation and cycli-
zation between dopamine (amine part) and 4-hydroxy-
phenylacetaldehyde (4-HPAA) (aldehyde part).2) This
reaction is the so-called Pictet–Spengler reaction, and
Materials and methods
Genetic techniques.
General DNA manipulations
were carried out as described by Sambrook and
Russell8) Plasmid DNA was purified from Escherichia
coli using a QIAprep Spin Miniprep kit (Qiagen,
Hilden, Germany). Restriction enzymes, a DNA liga-
tion kit, and DNA polymerase (PrimeStar® HS) were
purchased from Takara Bio (Shiga, Japan).
*Corresponding author. Email: Masakatsu_Nishihachijo@kn.kaneka.co.jp
Abbreviations: NCS, Norcoclaurine synthase; e.e., enantiomeric excess; 4-HPAA, 4-hydroxyphenylacetaldehyde.
© 2014 Japan Society for Bioscience, Biotechnology, and Agrochemistry

702
M. Nishihachijo et al.
Fig. 1. Biosynthetic pathway of benzylisoquinoline alkaloids in plants.
Notes: Norcoclaurine synthase (NCS) is involved in the first committed step of benzylisoquinoline alkaloid synthesis. It catalyzes the
condensation and cyclization between dopamine (amine part) and 4-hydroxyphenylacetaldehyde (aldehyde part), which is the so-called Enzymatic
Pictet–Spengler reaction.
Expression of recombinant CjNCS protein.
The
and Co., Franklin Lakes, NJ, USA). Ampicillin was
added to a final concentration of 0.1 mg/mL. The
recombinant cells were harvested by centrifugation,
suspended in 100 mM Tris-HCl buffer (pH 7.0), and
disrupted by sonication to obtain the cell-free extracts.
pUC19-derived expression vector pSNT-CjNCS was
constructed as follows. First, the sequence of the ribo-
some-binding site of pUCNT,9) which was developed by
Nanba et al. and carries the lac promoter and rrnBT1 ter-
minator for protein expression, was changed to TA-
AGGAGGTT from CACAGGAAAC by site-directed
mutagenesis. The resulting plasmid pSNT was digested
with NdeI and PstI, and ligated with a PCR-amplified,
NdeI- and PstI-digested DNA fragment containing the
CjNCS (CjPR10A) gene. The primers used were 5′-CC
ATATGCGTATGGAAGTGGTTCT-3′ (forward) and 5′-
TACTGCAGTTATTCGGAAGATTTGTG-3′ (reverse),
and the template used was a plasmid carrying the codon-
optimized CjNCS gene that was synthesized by Gen-
Script (Piscataway, NJ, USA). N-Terminally truncated
CjNCS variants were also constructed, in which the first
10, 19, 29, and 42 amino acid residues were removed.
The forward primers were 5′-TCCATATGCTGATGTT
CATTGGCACC-3′ (Δ10), 5′-TCCATATGGAACGCCT
GATTTTTAA-3′ (Δ19), 5′TCCATATGCTGCATCGCG
TTACGAAA-3′ (Δ29), and 5′-TCCATATGCACGAACT
GGAAGTGGCG-3′ (Δ42). The reverse primer was the
same as described above. The PCR-amplified DNA frag-
ments were sequenced by the dideoxy chain termination
method10) to ensure that no base changes other than
those planned had occurred.
Analytical methods.
The concentrations of dopa-
mine and isoquinoline alkaloids were determined using
high-performance liquid chromatography (HPLC)
equipped with a Finepak SIL C18T-5 (4.6 mm I.D. ×
250 mm) column (JASCO, Tokyo, Japan). Elution was
carried out by water/acetonitrile (6/4, by volume) con-
taining 3.2 g/L KH2PO4 and 1.5 g/L SDS (adjusted to
pH 3.5 with phosphoric acid) at a flow rate of 1 mL/
min and was monitored at 230 nm. The column was
kept at 30 °C. The optical purity of 6,7-dihydroxy-1-
propyl-1,2,3,4-tetrahydroisoquinoline and 6,7-dihy-
droxy-1-phenethyl-1,2,3,4-tetrahydroisoquinoline was
determined using an HPLC equipped with a Chiralpak
AD-H (4.6 mm I.D. × 250 mm) column (Daicel, Osaka,
Japan). The column was eluted by n-hexane/ethanol/
diethylamine (80/20/0.1, by volume) as the mobile
phase. The column temperature was kept at 30 °C, and
the elution was carried out at a flow rate of 1 mL/min
and monitored at 230 nm.
1H-nuclear magnetic resonance (NMR) spectra were
recorded in CD3OD or CDCl3 using a FT-NMR JNM-
ECA500 spectrometer (500 MHz; JEOL, Tokyo, Japan).
The content of CjNCS variants in the cell-free extract
The expression plasmid was introduced into E. coli
HB101, and the transformants were cultured at 20 or
30 °C for 24 h in 2 × YT medium (Becton, Dickinson

Pictet–Spengler reaction catalyzed by CjNCS
703
of E. coli was estimated by SDS-polyacrylamide gel
electrophoresis (10%), followed by Coomassie Brilliant
Blue staining.
acid (48%, 10.0 g, 59.3 mmol) to 6,7-dimethoxy-1-
phenethyl-1,2,3,4-tetrahydroisoquinoline (1.06 g, 3.37
mmol). The reaction mixture was stirred for 16 h at
90 °C, cooled to room temperature, and the pH was
adjusted to 7 with 7.5 M NaOH. Water (20 mL) and
ethyl acetate (10 mL) were then added to the mixture,
and it was stirred for 30 min at 5 °C to achieve crystal-
lization. The crystals were filtered out and washed with
water and dried at 50 °C in vacuo to give 520 mg of a
racemic product (1a) as a light brown solid. NMR δH
(CD3OD): 1.86(2H, m), 2.1–2.7(4H, m), 3.02(1H, m),
3.27 (1H, m), 3.5–4.4 (2H, brs), 3.91 (1H, m), 6.57
(1H, s), 6.60 (1H, s), 7.1–7.3 (5H, m).
Enzyme assays.
Catalytic activities of wild-type
and N-terminally truncated versions of CjNCS were
determined by measuring the consumption of dopa-
mine. The reaction mixture contained 100 mM Tris-
HCl buffer (pH 7.0), 10 mM dopamine-HCl, 10 mM
aldehyde, and 1% dimethyl sulfoxide (DMSO) in a
total volume of 1 mL. The reaction was initiated by
adding the enzyme, and the mixture was incubated at
30 °C for 15 min. The reaction was terminated by the
addition of the elution buffer used for the HPLC analy-
sis (see above), and the products were injected onto the
column. Spontaneous cyclization (non-enzymatic Pic-
tet–Spengler reaction) occurred under the test condi-
tions; therefore, the control experiment was carried out
in the absence of enzyme, and the amount of dopamine
consumed was subtracted. One unit of enzyme activity
was defined as the amount of enzyme that consumes
1 μmol of dopamine per min. The aldehyde part of the
substrate 3-(4-trifluoromethylphenyl)-1-propylaldehyde
(3) was prepared as described below. All the other
aldehydes used in this experiment were of analytical
grade and are commercially available.
Preparation of racemic 6,7-dihydroxy-1-propyl-
1,2,3,4-tetrahydroisoquinoline (2a).
Racemic (2a)
was prepared as described by Pesnot et al.13) Dopa-
mine-HCl (194 mg, 1.0 mmol) and 1-butylaldehyde
(111 mg, 1.5 mmol) were added to 1.94 mL of 100 mM
potassium phosphate buffer (pH 6.5) and stirred for 2 h
at 40 °C. The reaction products were then extracted
with 1-butanol (2 × 4 mL), and the extracts were dehy-
drated with sodium sulfate. The solvent was removed
in vacuo to give 241 mg racemic product (2a) as a yel-
low solid. NMR δH (CD3OD): 1.03 (3H, t), 1.51–1.54
(2H, m), 1.84–1.86 (1H, m), 1.97–2.00 (1H, m), 2.90
(1H, m), 2.96 (1H, m), 3.29–3.30 (1H, m), 3.46–3.50
(1H, m), 4.32–4.35 (1H, dd), 6.59 (1H, s), 6.64
(1H, s).
Preparation of 3-(4-trifluoromethylphenyl)-1-propy-
laldehyde (3).
3-(4-Trifluoromethylphenyl)-1-propa-
nol (90.6%, 8.27 g, 36.7 mmol) synthesized as
described by Sifferlen et al.11) was dissolved in 80 mL
of ethyl acetate and sodium hydrogen carbonate (9.38
g, 0.112 mol), NaBr (3.83 g, 37.2 mmol), 2,2,6,6-tetra-
methylpiperidine 1-oxyl (0.291 g, 1.86 mmol), and
water (80 mL) were added to the mixture at 0 °C.
Hypochlorous acid water (13.0%, 21.28 g, 37.2 mmol)
was then added dropwise, and the mixture was further
stirred for 1.5 h at 0 °C. The product was extracted with
ethyl acetate. The organic layer was washed with water
(2 × 40 mL) and concentrated in vacuo to give 8.08 g of
3-(4-trifluoromethylphenyl)-1-propylaldehyde as a light
orange oil. NMR δH (CDCl3): 2.81(2H, d, J = 7.3 Hz),
3.01(2H, t, J = 7.6 Hz), 7.30(2H, d, J = 8.0 Hz), 7.53
(2H, d, J = 8.1 Hz), 9.81(1H, s).
Enzymatic synthesis of optically active 6,7-dihy-
droxy-1-phenethyl-1,2,3,4-tetrahydroisoquinoline (1a)
and 6,7-dihydroxy-1-propyl-1,2,3,4-tetrahydroisoquino-
lines (2a).
The cell-free extract of E. coli strain
expressing CjNCS-Δ29 was incubated with 65.3 mM
dopamine-HCl (10 mg/mL dopamine) and hydrocinna-
maldehyde (added stepwise to a final 1.25 eq. of dopa-
mine) at 30 °C for 16 h (for 1a) or 1-butylaldehyde
(added stepwise to a final 2.9 eq. of dopamine) at
30 °C for 1.5 h (for 2a) in a total volume of 1 mL. In
the control experiment, the cell-free extract was omit-
ted. The reaction products were extracted with 1-buta-
nol and the organic layer was concentrated in vacuo.
The concentration and optical purity of the products
were determined using HPLC as described above.
Preparation of racemic 6,7-dihydroxy-1-phenethyl-
Results
1,2,3,4-tetrahydroisoquinoline (1a).
1-Phenethyl-6,7-
Expression of CjNCS in E. coli
dimethoxy-3,4-dihydroisoquinoline (1.48 g, 5 mmol)
synthesized as described by De Vries et al.12) was dis-
solved in 10 mL of methanol and stirred at 5 °C.
Sodium borohydride (189 mg, 5 mmol) was then added,
and the mixture was further stirred for 3 h at 25 °C.
The reaction mixture was poured slowly to cooled
water (10 mL) together with concentrated HCl (2.61 g,
25 mmol), and the pH was adjusted to 12 with 7.5 M
NaOH. After concentration in vacuo, the product was
extracted with ethyl acetate (20 mL). The organic layer
was washed with water (10 mL) and concentrated in
vacuo to give 1.58 g of 6,7-dimethoxy-1-phenethyl-
1,2,3,4-tetrahydroisoquinoline as a light yellow oil.
Demethylation was performed by adding hydrobromic
In a previous study, we employed the T7 system for
CjNCS (CjPR10A) expression; however, in this study,
we newly constructed a pUC-derived expression vector
because this expression system does not require an
expensive inducer such as isopropyl-β-1-thiogalactopy-
ranoside. This is important from the viewpoint of later
commercially adopting the process.
NCS is predicted to be localized in a vesicular com-
partment.5) SignalP server (http://www.cbs.dtu.dk/ser
vices/SignalP) predicted the presence of the signal
peptide (first 19 amino acid resides) in the CjNCS
sequence. The CjNCS protein possessing the signal
peptide was not efficiently expressed in E. coli as

704
M. Nishihachijo et al.
Fig. 2. SDS-polyacrylamide gel electrophoresis of the E. coli cell-free extracts expressing CjNCS variants.
Notes: N-Terminally truncated versions of CjNCS were constructed using a pUC-derived vector and expressed in E. coli. S: soluble fraction,
P: precipitated fraction.
expected, and no corresponding band appeared in
the SDS-polyacrylamide gel electrophoresis (Fig. 2).
Reflecting this, quite low activity (0.03 U/mL) was
detected in the cell-free extract when it was measured
using dopamine and hydrocinnamaldehyde as the sub-
strates. We then designed four N-terminally truncated
CjNCS variants to enhance the expression level in E.
coli. Among the tested constructs (Δ10, Δ19, Δ29, and
Δ42), CjNCS-Δ29 was found to be highly expressed as
a soluble form, although about half of the protein was
still recovered in an insoluble fraction. The ratio of the
soluble to insoluble fraction of recovered CjNCS-Δ29
did not change when the cells were cultivated at low
temperature (20 °C) and when the cells co-expressing
chaperones (GroES and GroEL) were used as an
expression host (data not shown). The activities of the
CjNCS variants apparently coincided with the amounts
of proteins recovered in the soluble fractions (Table 1).
The highest activity was obtained for CjNCS-Δ29 (1.
20 U/mL).
4-HPAA (physiological substrate) as the aldehyde part
(82−99% relative to phenylacetaldehyde). n-Alkyl
aldehydes were also efficient substrates (27−68%).
However, the enzyme is less active toward branched-
chain alkyl groups. These results support the idea that
CjNCS-Δ29 catalyzes the condensation and cyclization
reaction between dopamine and various alkyl and
aromatic aldehydes with considerable efficiency.
Preparative synthesis of optically active 1-substituted
tetrahydroisoquinolines
We synthesized two unnatural, 1-substituted tetrahy-
droisoquinolines (Table 3). Incubation of E. coli cell-
free extracts expressing CjNCS-Δ29 with dopamine
and hydrocinnamaldehyde yielded 19.0 g/L of 6,7-dihy-
droxy-1-phenethyl-1,2,3,4-tetrahydroisoquinoline (1a).
The molar yield and enantiomeric excess (e.e.) of 1a
were 86.0 and 95.3%, respectively (Fig. 3). In contrast,
in the control experiment without the enzyme, non-ster-
eoselective spontaneous condensation and cyclization
(non-enzymatic Pictet–Spengler reaction) between
dopamine and hydrocinnamaldehyde was observed. In
this case, the molar yield and e.e. were 10.6 and 9.4%,
respectively (Table 3).
Substrate specificity of CjNCS-Δ29
NCS requires two substrates to form a tetrahydroiso-
quinoline skeleton, one being amine and the other alde-
hyde. TfNCS was shown to strictly recognize the amine
part of the substrate (only the 3’-hydroxyphenyl-2-eth-
ylamine derivatives are accepted), while the recognition
of the aldehyde part is much more relaxed.14) We
examined the substrate specificity of CjNCS-Δ29 using
dopamine and various aldehydes as substrates (Table 2).
CjNCS-Δ29 showed activity toward aryl and aliphatic
aldehydes such as hydrocinnamaldehyde (2), 3-(4-tri-
fluoromethylphenyl)-1-propylaldehyde (3), 2-phenylpro-
pylaldehyde (4), 3-(4-isopropylphenyl)-isobutyraldehy
de (5), 1-butylaldehyde (6), isobutyraldehyde (7),
1-pentylaldehyde (8), 1-hexylaldehyde (9), 1-heputylal-
dehyde (10), and glutaraldehyde (11). Among these
non-physiological substrates, phenylacetaldehyde and
hydrocinnamaldehyde were almost equally accepted
and the most suitable. CjNCS-Δ29 efficiently accepts
aromatics with longer alkyl side chains in addition to
6,7-Dihydroxy-1-propyl-1,2,3,4-tetrahydroisoquino-
line (2a) (13.5 g/L) was also synthesized using the
Table 1. Activity of N-terminally truncated CjNCS.
Estimated from SDS-PAGEa
Constructs
Activityb (U/mL)
Soluble form
Insoluble form
CjNCS
–
–
+
+++
–
–
++++
++
+++
–
0.03
0.10
0.30
1.20
0.09
CjNCS-Δ10
CjNCS-Δ19
CjNCS-Δ29
CjNCS-Δ42
a“–” indicates that the corresponding protein band was not observed, and “+”
indicates the apparent quantity of the corresponding protein.
bActivity was measured using dopamine and hydrocinnamaldehyde as the
substrates.

Pictet–Spengler reaction catalyzed by CjNCS
Table 2. Aldehyde substrate specificity of CjNCS-Δ29.a
705
Entry
1
Substrate
Relative activity （%）
100
2
3
99
82
4
5
8
2
6
7
35
6
8
67
27
68
10
9
10
11
aActivity was determined by measuring the consumption of dopamine. In the
control experiment, the enzyme was omitted from the reaction mixture (see
Materials and Methods).
E. coli cell-free extracts expressing CjNCS-Δ29 from
dopamine and 1-butylaldehyde. The molar yield and
e.e. of 2a were determined to be 99.6 and 98.0%,
respectively. This reaction took only 1.5 h to complete.
The non-enzymatic reaction produced racemic 2a with
a molar yield and e.e. of 5 and 0.4%, respectively
(Table 3).
Discussion
Chemical methods for preparing optically active
1-substituted
tetrahydroisoquinolines
have
been
developed, but they are not very efficient.15−17) The
theoretical maximum yield of the optical resolution of

706
M. Nishihachijo et al.
Fig. 3. Chiral HPLC analysis of the reaction products catalyzed by CjNCS-Δ29.
Notes: Non-enzymatic (A) and enzymatic (B) Pictet–Spengler reactions catalyzed by NCS. Dopamine and hydrocinnamaldehyde were used as
the substrates. Chiralpak AD-H column was used as described in the Materials and Methods.
Scheme 1. Enzymatic synthesis of unnatural 1-substituted 1,2,3,4-tetrahydroisoquinolines.
a racemic compound is 50%. The asymmetric reduction
of 3,4-dihydroisoquinolines (prochiral imines) and the
chemical asymmetric Pictet–Spengler reaction require
expensive ligands and/or rare metals as catalysts or chi-
ral auxiliaries. In contrast, an enzymatic Pictet–Spen-
gler reaction catalyzed by NCS is capable of producing
optically active 1-substituted tetrahydroisoquinolines in
one step without using expensive reagents.
from the consumption of dopamine). It also accepted a
branched-chain alkyl group (α-substituted aldehydes,
entries 4, 5, and 7 in Table 2), although the activity
was significantly lower than that toward an n-alkyl
group. These results agree well with a recent report by
Pesnot et al., in which they showed using the intact,
His-tagged form of CjNCS that dopamine was con-
sumed when incubated with aliphatic and aromatic
aldehydes with a longer alkyl side chain. Isobutyralde-
hyde (7) was also recognized by the enzyme, albeit
poorly.20) On the contrary, Ruff et al. have reported that
the condensation product between dopamine and an α-
substituted aldehyde was not detectable when TfNCS
was used as a catalyst.4) This apparent discrepancy
might be caused by the difference in the enzyme source
(T. flavum or C. japonica) or by the difference in the
assay method (detection of dopamine consumption or
detection of product). Therefore, in order to confirm
whether CjNCS can accept α-substituted aldehydes as
substrates, it is necessary to isolate the condensation
product of α-substituted aldehydes and dopamine.
Functional expression of CjNCS-Δ29 in E. coli
From a practical viewpoint, it is important to prepare
“a catalyst” easily and cost-effectively. In this study,
we succeeded in overexpressing the CjNCS protein in
E. coli by deleting the first 29 amino acid residues.
The expression system did not require an expensive
inducer. The cell-free extract of E. coli expressing
CjNCS-Δ29 exhibited 40-fold higher activity than that
expressing intact CjNCS protein. Increase in the NCS
expression level and the rate of its recovery in a solu-
ble fraction by deleting the N-terminus has also been
observed for TfNCS, although the specific activity was
not compared between the two variants.18) Our results
suggest that the N-terminal 29 amino acid residues of
CjNCS are dispensable for its functional expression.
Interestingly, in the crystal structure of TfNCS, the first
39 amino acid residues of the B chain in the asymmet-
ric unit are not visible (PDB ID: 2VNE, 2VQ5).19)
Preparative synthesis of tetrahydroisoquinolines
Two optically active, 1-substituted 1,2,3,4-tetrahydro-
isoquinolines, 1a and 2a, were synthesized in high
optical purity and high yield. Although we did not
determine their steric configurations, the products are
likely the (S)-enantiomer according to the proposed cat-
alytic mechanism.20) Production of 1a is especially sig-
nificant because its derivatives are in high demand as
pharmaceutical intermediates.21)
Aldehyde substrate specificity of CjNCS
CjNCS-Δ29 efficiently recognized aromatic alde-
hydes and n-alkyl aldehydes as substrates (as judged

Pictet–Spengler reaction catalyzed by CjNCS
707
[cited 2011 May 24 ];2:326. Available from: http://www.nature.
com/ncomms/journal/v2/n5/full/ncomms1327.html
[8] Sambrook J, Russell DW. Molecular cloning, a laboratory man-
ual. 3rd ed. Cold Spring Harbor: Cold Spring Harbor Laboratory
Press; 2001.
[9] Nanba H, Ikenaka Y, Yamada Y, Yajima K, Takano M,
Takahashi S. Production of thermotolerant N-carbamyl-D-amino
acid amidohydrolase by recombinant Escherichia coli. J. Biosci.
Bioeng. 1999;87:149–154.
[10] Sanger F, Nicklen S, Coulson AR. DNA sequencing with chain-
terminating inhibitors. Proc. Natl. Acad. Sci. USA. 1977;74:5463–
5467.
[11] Sifferlen T, Koberstein R, Cottreel E, Boller A, Weller T,
Gatfield J, Brisbare-Roch C, Jenck F, Boss C. Structure-activity
relationship studies and sleep-promoting activity of novel
1-chloro-5,6,7,8-tetrahydroimidazo[1,5-a]pyrazine derivatives as
dual orexin receptor antagonists. Part 2. Bioorg. Med. Chem.
Lett. 2013;23:3857–3863.
[12] De Vries AHM, Domin D, Helms M, Imboden C, Koberstein R,
Nazir Z, Skranc W, Stanek M, Tschebull W, Verzijl GKM.
Actelion Pharmaceuticals Ltd. Trisubstituted 3,4-dihydro-1H-
isoquinolin compound, process for its preparation, and its use.
WO Patent 2009/083903. 2009 Jul. 9.
[13] Pesnot T, Gershater MC, Ward JM, Hailes HC. Phosphate medi-
ated biomimetic synthesis of tetrahydroisoquinoline alkaloids
Chem. Commun. 2011;47:3242–3244.
[14] Luk LY, Bunn S, Liscombe DK, Facchini PJ, Tanner ME.
Mechanistic studies on norcoclaurine synthase of benzylisoquin-
oline alkaloid biosynthesis: an enzymatic Pictet-Spengler reac-
tion Biochemistry. 2007;46:10153–10161.
Interestingly, despite the relative activity of CjNCS
toward butylaldehyde (Table 2, entry 6) being lower
than that for hydrocinnamaldehyde (entry 2), the yield
of 2a was higher than that of 1a from the preparative
synthesis. In addition, the reaction time required for the
synthesis of 2a was significantly shorter than that
required for 1a. We do not have a clear answer for this
discrepancy; however, it might be a result of the differ-
ent reaction conditions employed in the experiments,
including substrate concentrations, ratio of dopamine to
aldehyde, and the presence of DMSO.
Physiological implications
Phenethyl isoquinoline alkaloids such as dysoxyline
and colchicine are rare species of plant secondary
metabolites.22,23) In an earlier work, Battersby et al.
suggested that the synthetic pathway of these com-
pounds involves condensation between dopamine and
cinnamaldehyde24); however, cinnamaldehyde was not
found to be accepted by CjNCS.20) In this report, we
demonstrated that CjNCS can synthesize 1-phenethyl
tetrahydroisoquinoline from dopamine and hydrocinna-
maldehyde. These results suggest that plants may use
hydrocinnamaldehyde to synthesize phenethyl isoquino-
line alkaloids via the involvement of NCS.
In summary, we succeeded in developing an efficient
method for practically (industrially) producing unnatu-
ral, optically active 1-substituted tetrahydroisoquino-
lines, which should serve as useful chemicals for drug
discovery purposes.
[15] Vilhanová B, Matoušek V, Václavík J, Syslová K, Přech J,
Pecháčeka J, Šota P, Januščáka J, Tomanb J, Zápalc J, Kuzmac M,
Kačer P. Two optimized synthetic pathways toward a chiral pre-
cursor of Mivacurium chloride and other skeletal muscle relaxants.
Tetrahedron: Asymmetry. 2013;24:50–55.
[16] Noyori R, Ohta M, Hsiao Y, Kitamura M, Ohta T, Takaya H.
Asymmetric synthesis of isoquinoline alkaloids by homogeneous
catalysis. J. Am. Chem. Soc. 1986;108:7117–7119.
Acknowledgments
[17] Comins DL, Badawi MM. Asymmetric pictet-spengler synthesis
of tetrahydroisoquinolines. An enantioselective synthesis of
(−)-laudanosine. Tetrahedron Lett. 1991;32:2995–2996.
[18] Berkner H, Engelhorn J, Liscombe KD, Schweimer K, Wöhrl
BM, Facchini PJ, Rösch P, Matecko I. High-yield expression
and purification of isotopically labeled norcoclaurine synthase, a
Bet v 1-homologous enzyme, from Thalictrum flavum for NMR
studies. Protein Expr. Purif. 2007;56:197–204.
We wish to thank Dr. Masahiko Yamada (Biochemi-
cal and Medical Business Development Division, Kane-
ka Corporation) for coordinating this work. The
employment of T.K. is supported by the Institute for
Fermentation, Osaka.
[19] Ilari A, Franceschini S, Bonamore A, Arenghi F, Botta B,
Macone A, Pasquo A, Bellucci L, Boffi A. Structural basis
of enzymatic (S)-norcoclaurine biosynthesis. J. Biol. Chem.
2009;284:897–904.
[20] Pesnot T, Gershater CM, Ward MJ, Hailes CH. The catalytic
potential of Coptis japonica NCS2 revealed – development and
utilisation of a fluorescamine-based assay. Adv. Synth. Catal.
2012;354:2997–3008.
[21] Brisbare-Roch C, Dingemanse J, Koberstein R, Hoever P,
Aissaoui H, Flores S, Mueller C, Nayler O, van Gerven J,
de Haas SL, Hess P, Qiu C, Buchmann S, Scherz M, Weller T,
Fischli W, Clozel M, Jenck F. Promotion of sleep by targeting
the orexin system in rats, dogs and humans. Nat. Med.
2007;13:150–155.
[22] Aladesanmi AJ, Ilesanmi OR. Phytochemical and pharmacologi-
cal investigation of the cardioactive constituents of the leaf of
Dysoxylum lenticellare. J. Nat. Prod. 1987;50:1041–1044.
[23] Maier UH, Zenk MH. Colchicine is formed by para-para phenol
coupling from autumnaline. Tetrahedron Lett. 1997;38:7357–
7360.
References
[1] Stadler R, Kutchan TM, Loeffler S, Nagakura N, Cassels B,
Zenk MH. Revision of the early steps of reticuline biosynthesis.
Tetrahedron Lett. 1987;28:1251–1254.
[2] Samanani N, Facchini PJ. Isolation and partial characterization
of norcoclaurine synthase, the first committed step in benzyliso-
quinoline alkaloid biosynthesis, from opium poppy. Planta.
2001;213:898–906.
[3] Bonamore A, Rovardi I, Gasparrini F, Baiocco P, Barba M, Molina-
ro C, Botta B, Boffi A, Macone A. An enzymatic, stereoselective
synthesis of (S)-norcoclaurine. Green Chem. 2010;12:1623–1627.
[4] Ruff BM, Bräseb S, O’Connor SE. Biocatalytic production of
tetrahydroisoquinolines. Tetrahedron Lett. 2012;53:1071–1074.
[5] Minami H, Dubouzet E, Iwasa K, Sato F. Functional analysis of
norcoclaurine synthase in Coptis japonica. J. Biol. Chem.
2007;282:6274–6282.
[6] Minami H, Kim JS, Ikezawa N, Takemura T, Katayama T, Kuma-
gai H, Sato F. Microbial production of plant benzylisoquinoline
alkaloids. Proc. Natl. Acad. Sci. USA. 2008;105:7393–7398.
[7] Nakagawa A, Minami H, Kim JS, Koyanagi T, Katayama T,
[24] Battersby AR, Bink R, Reynolds JJ, Yeowell DA. Alkaloid bio-
synthesis. Part VI. The biosynthesis of colchicine. J. Chem. Soc.
1964; 4257–4268.
Sato F, Kumagai H.
production of plant alkaloids. Nature Commun. [Internet]. 2011
A bacterial platform for fermentative