Phosphate mediated biomimetic synthesis of tetrahydroisoquinoline alkaloids

Article abstract of DOI:10.1039/c0cc05282e

A one-pot synthesis of tetrahydroisoquinoline alkaloids in a phosphate buffer has been achieved, and a reaction mechanism proposed. The utilisation of mild reaction conditions readily afforded a range of isoquinolines, including norcoclaurine.

Full text of DOI:10.1039/c0cc05282e

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
ChemComm
Cite this: Chem. Commun., 2011, 47, 3242–3244
www.rsc.org/chemcomm
COMMUNICATION
Phosphate mediated biomimetic synthesis of tetrahydroisoquinoline
alkaloidsw
Thomas Pesnot,a Markus C. Gershater,b John M. Wardb and Helen C. Hailes*a
Received 30th November 2010, Accepted 10th January 2011
DOI: 10.1039/c0cc05282e
A one-pot synthesis of tetrahydroisoquinoline alkaloids in a
phosphate buffer has been achieved, and a reaction mechanism
proposed. The utilisation of mild reaction conditions readily
afforded a range of isoquinolines, including norcoclaurine.
The benzylisoquinoline alkaloids (BIAs) form one of the most
varied groups of natural products found in plants and mammals.1
BIAs, one of the major classes of tetrahydroisoquinolines, have
been identified as pharmacologically active, showing antitumor,2
anti-microbial,2,3 anti-inflammatory,4 anti-HIV,5 and analgesic
activities.6 While naturally occurring tetrahydroisoquinoline
alkaloids can be obtained by plant extraction methods, the yield
is alkaloid dependent and most compounds are found in very
small quantities or as part of a complex mixture. More recently
metabolic engineering strategies have started to be investigated
for the generation of larger quantities of alkaloids.7 In plants
norcoclaurine synthases (NCSs), catalyse the first committed step
in the BIA biosynthetic pathway, between dopamine 1 and
4-hydroxyphenylacetaldehyde (4-HPAA) 2.8 However, studies
to date have demonstrated that native NCSs exert a poor
tolerance for non-natural substrate analogues,9 and therefore at
present for greater structural diversity chemical synthesis is
required. A key synthetic approach to tetrahydroisoquinolines
is the Pictet–Spengler reaction which originally described the
reaction between b-phenethylamine and formaldehyde dimethyl
acetal with hydrochloric acid.10 In general, less harsh conditions
are required with indoles than phenethylamines, where high
temperatures and strong acids or superacids are normally needed,
which are incompatible with poorly stable aldehydes or amines.11
More recently, the coupling of dopamine and L-DOPA with
glyceraldehyde and glucose under physiologically relevant
(pH 7.4 phosphate buffer) and rigorous anaerobic conditions
have been described, although the reasons for the enhancement of
the Pictet–Spengler reaction were not rationalised.12 However, to
prepare BIAs alternative chemical strategies are used such as the
multi-step Bischler–Napieralski cyclisation and dehydration
procedure, together with protecting group manipulations.13
Scheme 1 Formation of 3. Reaction conditions: (i) 1 : 1 co-solvent/
buffer (0.1 M), 37–50 1C (Table 1).
For this reason new concise but versatile synthetic methodologies
are highly sought after, indeed recently procedures using a calcium-
based Lewis acid catalyst and o-benzenedisulfonamide as a
reusable acid catalyst have been reported.14 Here, we report a high
yielding biomimetic reaction for the synthesis of BIAs and analo-
gues from variously decorated phenethylamines and aldehydes.
With the aim of developing facile Pictet–Spengler reaction
conditions the coupling of dopamine 1 and 4-HPAA 2 to give
norcoclaurine 3 was investigated (Scheme 1). The main synthetic
challenge in this reaction is the poor stability of both substrates:
dopamine readily undergoes oxidative degradation, while
4-HPAA polymerises under both basic and acidic conditions.
4-HPAA 2 was synthesised using the previously reported
Parikh–Doering oxidation of 2-(4-hydroxyphenyl)ethanol.15
Initially the synthesis of 3 was investigated using 1ꢀHCl (4 mM)
and 2 (4.8 mM) in a 1 : 1 mixture of acetonitrile (to aid full
substrate solubility) and potassium phosphate buffer (KPi, 0.1 mM)
at physiological pH and temperature. This yielded norcoclaurine 3
in limited quantities (5% after 1 h). This was interesting since some
publications where NCSs are used have referred to a minor
background chemical reaction, although others use KPi as buffer
and similar reaction conditions with no reported non-enzymatic
reaction.8d,9,16 Further studies established that in acetonitrile/KPi
at pH o 4 and pH 4 8 no 3 was formed, and under alkaline
conditions 4-HPAA 2 underwent rapid degradation. At inter-
mediate pHs the reaction proceeded smoothly, and yields reached
40% at pH 6 after 1 h. The conversion rate was further increased
to 77% by elevating the reaction temperature to 50 1C: higher
temperatures lead to the degradation of both 1 and 2. These results
were intriguing since the condensation usually requires strong acid
catalysis, but our data suggested that the buffer might be a key
component in the reaction, acting as an effective but mild catalyst.
To verify this hypothesis the synthesis of norcoclaurine was carried
out in various buffers including HEPES, Tris, MOPS and different
phosphates. The results unequivocally showed that phosphates
promoted the reaction (Scheme 1 and Table 1).
a Department of Chemistry, University College London,
20 Gordon Street, London, WC1H 0AJ, UK.
E-mail: h.c.hailes@ucl.ac.uk; Fax: +44 (0)20 7679 7463;
Tel: +44 (0)20 7679 7463
b Research Department of Structural & Molecular Biology,
University College London, Gower Street, London WC1E 6BT, UK
w Electronic supplementary information (ESI) available: Character-
isation data for 3, 4, 6 and 8. See DOI: 10.1039/c0cc05282e
c
3242 Chem. Commun., 2011, 47, 3242–3244
This journal is The Royal Society of Chemistry 2011

View Article Online
Table 1 Influence of buffer on the formation of 3 (Scheme 1)
Table
5a–5g
3
Reaction of phenylacetaldehyde with phenethylamines
Buffer
Conversion (%)
Buffer
Conversion (%)
Tris
o1
o1
o1
o1
2
KH2PO4
NaH2PO4
UMP
Glc-1-P
Na4P2O7
Water
77
75
75
74
45
o1
HEPES
B(OH)3
Na3VO4
KHCO3
KHSO4
Amine
Product
Yield (%)
4
1, R1 = OH, R2 = OH, R3 = H
5a, R1 = OH, R2 = OH, R3 = OH
5b, R1 = H, R2 = OH, R3 = H
5c, R1 = OH, R2 = H, R3 = H
5d, R1 = OMe, R2 = OMe, R3 = H
5e, R1 = H, R2 = H, R3 = H
5f, R1 = NO2, R2 = H, R3 = H
5g, R1 = NH2, R2 = H, R3 = H
4k
6a
6b
6c
6d
6e
6f
73
9
0
74
0
0
Reaction conditions: 1ꢀHCl (4 mM), 2 (4.8 mM), 1 : 1 CH3CN/buffer
(0.1 M), pH 6, 50 1C, 1 h. Conversions determined by HPLC.
Potassium or sodium salts of phosphate, uridine-50-mono-
phosphate (UMP) as well as glucose-1-phosphate (Glc-1-P) all
acted as efficient catalysts. Other buffers, whether carbonate, sulfate
or amines, had a negligible effect. Borate and vanadate have been
reported to act as phosphate mimics, but no 3 was generated.17,18
Taken together, these results suggested that the phosphate moiety
was essential for the observed biomimetic synthesis of norcoclaurine.
It was demonstrated that the reaction also displayed a large
tolerance for co-solvents (see ESIw), where the highest conversions
were achieved using acetonitrile, methanol and DMSO.
0
52
6g
Reaction conditions: 1ꢀHCl or 5a–5g (1 equiv.), aldehyde (1.2 equiv.),
1 : 1 CH3CN/KPi (0.1 M) pH 6, 50 1C, 12 h.
We anticipate that reactions giving products in lower yields could
be optimised using alternative co-solvents, since the smaller
quantities of product generated probably reflected poor substrate
solubility in the solvent mixture used.
The general applicability of the reaction was then explored
for the production of variously decorated isoquinolines.
DopamineꢀHCl was reacted with a set of aldehydes (Table 2).
Aliphatic as well as aromatic and heteroaromatic aldehydes
readily reacted to generate the tetrahydroisoquinolines
(4a–b, 4c–k and 4l–n, respectively) in 31% to 88% isolated
yields. For example, phenylacetaldehyde gave 14-deoxy-norco-
claurine 4k in 73% yield. These results demonstrated that the
reaction is highly versatile and provides an attractive tool for
the rapid synthesis of naturally occurring (e.g. norcoclaurine)
and non-natural isoquinoline alkaloids.
In sharp contrast to the aldehydes, and consistent with
previous studies on the Pictet–Spengler condensation,19 a limited
array of phenethylamines acted as substrates for the reaction
(Table 3). Although dopamine 1, 5-hydroxydopamine 5a, and
3-hydroxyphenethylamine 5c all yielded the corresponding
isoquinoline products with phenylacetaldehyde to give 4k, 6a
and 6c, in the absence of the 3-hydroxyl functionality in regio-
isomer 5b, and analogues 5d–5f the reaction did not proceed.
Interestingly, when a 3-methoxy group was present in 5d this also
precluded the formation of product. However, the presence of a
3-amino group in 5g readily gave the isoquinoline 6g in 52%
yield. Taken together, these results suggested that an available
hydroxyl or amine electron donating group meta to the
ethylamine substituent is required for cyclisation. In addition,
these reactions were highly regioselective with cyclisation at the
least hindered para position, relative to the amine and phenolic
groups, and consistent with a cyclisation mechanism involving
direct formation of the 6-membered isoquinoline, rather than a
5-membered spirocyclic intermediate.9,11a
The reaction did not require any protection of phenolic
functionalities, which could easily be used in subsequent
derivatisation to rapidly generate still greater structural diversity.
The reaction also proved attractive for the synthesis of a
deuterated analogue 4k-2H2, simply prepared by pre-treatment
of the aldehyde in D2O. This together with the synthesis of the
quinolinyl and ferrocenyl derivatives (4n and 4o, respectively)
highlighted the utility of the reaction in rapidly preparing labelled
(including radiolabelled), fluorescent, or electrochemical probes.
In order to explore the synthesis of stereochemically
enriched isoquinoline alkaloids, analogous reactions were
performed using single isomer amines 7a and 7b or aldehydes
(Table 4). All reactions again proceeded well giving 8a–8d in
43% to 94% yield, further demonstrating the versatility of the
reaction. Stereoselectivities were observed for the reaction
between amine 7a and phenylacetaldehyde, and 1ꢀHCl and
(R)-(+)-2,2-dimethyl-1,3-dioxolane-4-carboxaldehyde to give
8a and 8c (deprotection of the diol from its isopropylidene
ketal occurred during the reaction), respectively. The highest
diastereomeric ratio (dr) (1 : 2.6) was observed in tetraol 8c,
where the major product formed (1R,10S) was consistent with the
Felkin-Ahn model of asymmetric induction, as previously.12d
Substrates with more remote stereocentres not surprisingly
gave lower stereoselectivities, although with norepinephrine
(7a) a dr of 1 : 1.6 was observed in 8a. This may have been due
to phosphate complexation at C4-OH, reducing access to one
face of the imine. Three of the four diastereoisomers
Table 2 Reaction of dopamine with selected aldehydes to give 4a–4o
Yield
Product (%)
Yield
Product (%)
R
R
4-HOC6H4CH2
Me
Cyclopropyl
Ph
4-O2NC6H4
4-MeOC6H4
4-HOC6H4
3-HOC6H4
2-HOC6H4
3
63
75
60
75
51
31
35
80
86
4-BrC6H4
4-FC6H4
PhCH2
4i
4j
4k
47
41
73
4a
4b
4c
4d
4e
4f
4g
4h
PhCD2 (90%) 4k-2H2 83
2-Thienyl
4l
79
77
88
44
3-Pyridinyl
3-Quinolinyl
Ferrocenyl
4m
4n
4o
Reaction conditions: 1ꢀHCl (1 equiv.), aldehyde (1.2 equiv.), 1 : 1
CH3CN/KPi (0.1 M) pH 6, 50 1C, 12 h for larger scale reactions (see ESIw).
c
This journal is The Royal Society of Chemistry 2011
Chem. Commun., 2011, 47, 3242–3244 3243

View Article Online
Table 4 Stereoselective synthesis of tetrahydroisoquinolines
catalysts for the reaction which shows high tolerance for the
aldehyde component. This one-step methodology involves
mild reaction conditions, and is a versatile alternative to
previously reported chemical routes to tetrahydroisoquinoline
alkaloids. In addition, the concomitant use of single isomer
substrates allowed the preparation of alkaloid diastereo-
isomers. Finally, a mechanism highlighting the essential role
of phosphate has been proposed. This reaction is one example
of the essential role phosphate can play as a catalyst and
may have had an evolutionary role in isoquinoline alkaloid
biosynthetic pathways.
Amine
Aldehyde
Product dr
Yield
7a R1 = OH, R3 = PhCH2
R2 = H
8a 1 : 1.6
27% (1S,4R)a
43% (1R,4R)a
30% (1R,3S)
35% (1S,3S)
26% (1S,10S)
68% (1R,10S)
7b R1 = H, R3 = PhCH2
R2 = CO2H
8b 1 : 1.2
1ꢀHCl R1 =
8c 1 : 2.6
(deprotected
tetraol)
R3 =
We thank the BBSRC (BB/G014426/1) for funding T. P.
and M. C. G.
R2 = H
1ꢀHCl R1 =
8d 1 : 1
43%
(not resolved)
R3 =
R2 = H
Notes and references
Reaction conditions: Amine (1 equiv.), aldehyde (1.2 equiv.), 1 : 1
CH3CN/KPi (0.1 M) pH 6, 50 1C, 12 h.a Tentatively assigned (see ESIw).
1 K. W. Bentley, Nat. Prod. Rep., 2006, 23, 444.
2 J. D. Scott and R. M. Williams, Chem. Rev., 2002, 102, 1669.
3 F. R. Stermitz, P. Lorenz, J. N. Tawara, L. A. Zenewicz and
K. Lewis, Proc. Natl. Acad. Sci. U. S. A., 2000, 97, 1433.
4 J. Cortijo, V. Villagrasa, R. Pons, L. Berto, M. Marti-Cabrera,
M. Martinez-Losa, T. Domenech, J. Beleta and E. J. Morcillo,
Br. J. Pharmacol., 1999, 127, 1641.
5 Y. Kashiwada, A. Aoshima, Y. Ikeshiro, Y.-P. Chen,
H. Furukawa, M. Itoigawa, T. Fujioka, K. Mihashi,
L. M. Cosentino, S. L. Morris-Natschke and K.-H. Lee, Bioorg.
Med. Chem., 2005, 13, 443.
6 A. J. Goodman, B. Le Bourdonnec and R. E. Dolle, ChemMedChem,
2007, 2, 1552.
7 E. Leonard, W. Runguphan, S. O’Connor and K. J. Prather,
Nat. Chem. Biol., 2009, 5, 292.
8 (a) A. Bonamore, M. Barba, B. Botta, A. Boffi and A. Macone,
Molecules, 2010, 15, 2070; (b) A. Ilari, S. Franceschini, A. Bonamore,
F. Arenghi, B. Botta, A. Macone, A. Pasquo, L. Bellucci and A. Boffi,
J. Biol. Chem., 2009, 284, 897; (c) H. Berkner, K. Schweimer,
Scheme
2 Proposed phosphate mediated mechanism for the
Pictet–Spengler reaction. Pi is inorganic phosphate.
I. Matecko and P. Rosch, Biochem. J., 2008, 413, 281;
¨
(d) N. Samanani and P. J. Facchini, Planta, 2001, 213, 898.
9 L. Y. P. Luk, S. Bunn, D. K. Liscombe, P. J. Facchini and
M. E. Tanner, Biochemistry, 2007, 46, 10153.
synthesised could be resolved by preparative HPLC, and
provided rapid access to the single isomer products.
10 A. Pictet and T. Spengler, Chem. Ber., 1911, 44, 2030.
11 (a) R. Quevedo, E. Baquero and M. Rodiguez, Tetrahedron Lett.,
2010, 51, 1774; (b) S. W. Youn, Org. Prep. Proced. Int., 2006, 38, 505;
(c) A. Yokoyama, T. Ohwada and K. Shudo, J. Org. Chem., 1999, 64,
611; (d) E. D. Cox and J. M. Cook, Chem. Rev., 1995, 95, 1797;
Overall, the results observed suggest that phosphate plays an
essential role in this Pictet–Spengler cyclisation (Scheme 2). The
first step will involve formation of the iminium intermediate at pH
6. A phosphate anion or dianion can then undergo nucleophilic
attack onto the iminium intermediate, to produce a highly reactive
aminophosphate A, in equilibrium with the iminium species.
Phosphate can then act as a base, deprotonating C6-OH (or
C6-+NH3 in 5g) and activating the ring for addition to the
imine at the para-position, or the aminophosphate A, resulting
in cyclisation. This would explain the non-reactivity of 5b and
5d–5f in the reaction. Rearomatisation could finally proceed
by a phosphate mediated intra- (path a) or intermolecular
(path b) abstraction of the 8a-H proton. In previous work by
Sutherland and co-workers the role of phosphate in prebiotic
chemistry and nucleophilic addition to cyanamide has been
proposed, as well as acting as a general base catalyst and
buffer.20 The uniqueness of phosphate in catalysing the
reaction here is believed to result from its basicity (abstraction
of C6-OH and 8a-H protons) and potential intermediacy of
aminophosphate A and ability to form a six-membered
intermediate (path a) facilitating rearomatisation.
(e) C. Schopf and H. Bayerle, Liebigs Ann. Chem., 1934, 513, 190.
¨
12 (a) P. Manini, M. d’Ischia and G. Prota, J. Org. Chem., 2001, 66,
5048; (b) P. Manini, M. d’Ischia and G. Prota, Bioorg. Med. Chem.,
2001, 9, 923; (c) P. Manini, M. d’Ischia and G. Prota, J. Org. Chem.,
2000, 65, 4269; (d) P. Manini, M. d’Ischia, R. Lanzetta, M. Parrilli
and G. Prota, Bioorg. Med. Chem., 1999, 7, 2525.
13 (a) M. Chrzanowska and M. D. Rozwadowska, Chem. Rev., 2004,
104, 3341; (b) M. L. Mohler, G.-S. Kang, S.-S. Hong, R. Patil,
O. V. Kirichenko, W. Li, I. M. Rakov, E. E. Geisert and
D. D. Miller, J. Med. Chem., 2006, 49, 5845.
14 (a) M. J. Vanden Eyden, K. Kunchithapatham and J. P. Stambuli,
J. Org. Chem., 2010, 75, 8542; (b) M. Barbero, S. Bazzi,
S. Cadamuro and S. Dughera, Tetrahedron Lett., 2010, 51, 6356.
15 T. Hirose, T. Sunazuka, T. Zhi-Ming, M. Handa, R. Uchida,
K. Shiomi, Y. Harigaya and S. Omura, Heterocycles, 2000, 53, 777.
16 (a) H. Minami, E. Dubouzet, K. Iwasa and F. Sato, J. Biol. Chem.,
2007, 282, 6274; (b) M. Rueffer, H. El-Shagi, N. Nagakura and
M. H. Zenk, FEBS Lett., 1981, 129, 5.
17 M. Sugiyama, Z. Hong, L. J. Whalen, W. A. Greenberg and
C. H. Wong, Adv. Synth. Catal., 2006, 348, 2555.
18 G. Huyer, S. Liu, J. Kelly, J. Moffat, P. Payette, B. Kennedy,
G. Tsaprailis, M. J. Gresser and C. Ramachandran, J. Biol. Chem.,
1997, 272, 843.
To summarise, reaction conditions have been identified for
the phosphate mediated Pictet–Spengler cyclisation of
phenethylamines with aldehydes, generating structurally
diverse tetrahydroisoquinolines. Phosphates are excellent
19 K. Mothes and H. R. Schutte, Angew. Chem., Int. Ed. Engl., 1963,
2, 441.
¨
20 M. W. Powner, B. Gerland and J. D. Sutherland, Nature, 2009,
459, 239.
c
3244 Chem. Commun., 2011, 47, 3242–3244
This journal is The Royal Society of Chemistry 2011