Biogenetic hypothesis and first steps towards a biomimetic synthesis of haouamines

Article abstract of DOI:10.1039/b613737g

A detailed hypothesis for the biogenesis of haouamines is reported herein, supported by experiments headed towards biomimetic synthesis of these compounds. The Royal Society of Chemistry.

Full text of DOI:10.1039/b613737g

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Biogenetic hypothesis and first steps towards a biomimetic synthesis of
haouamines{
Edmond Gravel, Erwan Poupon* and Reynald Hocquemiller
Received (in Cambridge, UK) 21st September 2006, Accepted 4th December 2006
First published as an Advance Article on the web 19th December 2006
DOI: 10.1039/b613737g
center ring suggests that the biogenetic origin is likely to be an
assembly of simple L-phenylalanine derived units which can be
traced four times in the final structure of 1. Baran and colleagues
suggested a condensation of four meta-hydroxylated phenylace-
taldehyde molecules with ammonia as the nitrogen source followed
by oxidative events. They also reported they had failed to take
advantage of that route. We suggest a modified hypothesis.
Central in our proposals is the formation of the tetrahydropyridine
core through a natural Chichibabin-like pyridine synthesis reaction
involving three aldehydes and a primary amine as the nitrogen
source, all units being presumably derived from the aforemen-
tioned amino-acid.
A detailed hypothesis for the biogenesis of haouamines is
reported herein, supported by experiments headed towards
biomimetic synthesis of these compounds.
Marine organisms belonging to the Ascidiacea class are a rich
source of structurally unique secondary metabolites. Among the
most remarkable of these, Haouamines A 1 and B were isolated by
E. Zub´ıa and co-workers in 2002 from Aplidium haouarianum
collected off Tarifa Island (Spain).1 Haouamines A and B belong
to a new family of complex polycyclic alkaloids (Fig. 1).
The molecular complexity of 1 constitutes an intellectual
challenge for organic chemists. In fact, several groups have
disclosed their own approaches to the construction of haoua-
mines.2–4 These efforts culminated in a total synthesis by the Baran
team in early 2006,5 soon followed by a formal total synthesis by
Weinreb and co-workers.6 We wish to report in this communica-
tion the first biomimetically inspired approach to haouamine A
featuring an expeditious assembly of an advanced intermediate
towards the total synthesis of 1.
Based on the fact that, in nature, amines can be transformed
into aldehydes through an oxidative deamination process, we can
assume that from L-phenylalanine, through hydroxylation to
L-m-tyrosine7,8 and decarboxylation, amine 2 can be obtained and
then undergo oxidative deamination to give rise to meta-
hydroxylated phenylacetaldehyde 3 (Scheme 1). The condensation
of 2 and three molecules of 3 will lead to the formation of
dihydropyridinium 4 (that already contains all atoms of the final
molecule) which can then be reduced to compound 5. The
formation of the C8–C9 bond can be seen as the result of an ortho–
para coupling9 that occurs under oxidative conditions{ to yield
intermediate 6. Such conditions might as well lead to the oxidation
of C26 and generate an allylic hydroxyle group (7). It has been
reported by Rawal and co-workers that under acidic conditions,
the remaining bond (C24–C26) is then easily formed, via
carbocation formation,2 completing the biosynthesis of haouamine
A 1 (Scheme 2).
The combination of ring systems into a single and small
architectural unit like haouamine A leads one to wonder how such
structures can be assembled in nature. The tetrahydropyridine
Biomimetic investigations were performed with this possible
biosynthetic route taken into consideration, and it was decided to
study the outcome of the condensation reaction of 3-methoxy-
phenethylamine 8 with 3-methoxyphenylacetaldehyde 9. The two
species were reacted together (in a 1 : 3 ratio) along with catalytic
quantities of ytterbium triflate in different solvents, at room
temperature. After 20 hours, pyridinium 11§ was found as the
major product, whatever the solvent was, with yields of 60 to 75%
Fig. 1 Structures of haouamines.
Laboratoire de Pharmacognosie associe´ au CNRS (UMR 8076 –
´
BioCIS), Centre d’Etudes Pharmaceutiques, Universite´ Paris-Sud 11, 5,
rue Jean-Baptiste Cle´ment, 92290, Chaˆtenay-Malabry, France.
E-mail: erwan.poupon@cep.u-psud.fr; Fax: +33 1 46 83 55 99;
Tel: +33 1 46 83 55 86
{ Electronic supplementary information (ESI) available: Experimental
procedures as well as copies of 1H and 13C NMR spectra are provided for
compounds 11, 12 and 14. See DOI: 10.1039/b613737g
Scheme 1 Formation of precursors 2 and 3 from L-phenylalanine.
Chem. Commun., 2007, 719–721 | 719
This journal is ß The Royal Society of Chemistry 2007

View Article Online
Scheme 2 Proposed biogenetic hypothesis for haouamine A 1.
(Scheme 3). The reaction involves a cascade of reactions similar to
the ones depicted on Scheme 2 (up to product 4). Results highlight
the great interest of rare earth-metal triflate catalyzed reactions in
organic synthesis: the reaction takes place in very mild conditions
instead of high temperatures and/or pressures historically required
for the Chichibabin pyridine synthesis reaction.10 Ytterbium
triflate plays the role of a Lewis acid11 and probably activates
both aldehyde and imine groups of the reactants and intermedi-
ates. It should be noted that no traces of dihydropyridinium salt 10
(the logical outcome of the reaction) were detected in the crude
reaction mixture. Spontaneous oxidations of dihydropyridinium
salts into pyridinium salts have previously been observed
with Chichibabin-like reactions,12 and appear to be especially
prominent when condensations are carried out with
phenylacetaldehydes.
Compound 11 was then easily transformed into 12" by reaction
with tribromoborane (4.1 equiv.) in methylene chloride. The
addition was performed at 278 uC and the mixture was allowed to
slowly warm up to room temperature within the next three hours.
Compound 12 was obtained in very good yield without any
column purification needed.
Our initial plan was to perform reduction of 11 with sodium
borohydride but the reaction only gave very limited quantities of
the desired product 14I as a mixture of cis/trans isomers
Scheme 3 Biomimetic self-assembly of pyridinium 11 followed by
phenol deprotection to yield compound 12.
Scheme 4 Reduction of 11 into tetrahydropyridine 14.
720 | Chem. Commun., 2007, 719–721
This journal is ß The Royal Society of Chemistry 2007

View Article Online
141.3, 141.4, 159.8; MS (ES) m/z 550 ([M + H]+), 444 (100); HRMS (ES)
(Scheme 4). The reason for such low yields is yet under
investigation in our laboratory but it is likely to be an issue of
pseudo-dimerisations through Diels–Alder type cycloadditions
involving 14 and its direct precursor 13.13
calcd for C36H39NO4H+: 550.2919, found: 550.2921.
1 L. Garrido, E. Zub´ıa, M. J. Ortega and J. Salva´, J. Org. Chem., 2003,
68, 293.
With compound 14 obtained in too little amounts, it was
decided to try out different types of oxidative conditions applied to
product 12 in order to see if the desired C8–C9 coupling can take
place despite high constraint resulting from the pyridinium’s
planarity. The outcome of such reactions is still under investigation
in our laboratory.
2 N. D. Smith, J. Hayashida and V. H. Rawal, Org. Lett., 2005, 7, 4309.
3 M. A. Grundl and D. Trauner, Org. Lett., 2006, 8, 23.
4 P. Wipf and M. Furegati, Org. Lett., 2006, 8, 1901.
5 P. S. Baran and N. Z. Burns, J. Am. Chem. Soc., 2006, 128, 3908.
6 J. H. Jeong and S. M. Weinreb, Org. Lett., 2006, 8, 2309.
7 It is known that the oxidation of L-phenylalanine to L-meta-tyrosine
mostly results from the attack of hydroxyl free radical: H. Kaur,
I. Fagerheim, M. Grootveld, A. Puppo and B. Halliwell, Anal.
Biochem., 1988, 172, 360.It has also been reported that tyrosine-
hydroxylase may produce small amounts of L-meta-tyrosine (as well as
L-para-tyrosine and L-DOPA of course) from L-phenylalanine:
M. H. Fukami, J. Haavik and T. Flatmark, Biochem. J., 1990, 268,
525; J. H. Tong, A. D’Iorio and N. L. Benoiton, Biochem. Biophys. Res.
Commun., 1971, 44, 229.
8 The peculiar L-meta-tyrosine pattern has been described in very few
natural products such as pacidamycins: R. H. Chen, A. M. Buko,
D. N. Whittern and J. B. McAlpine, J. Antibiot., 1993, 42, 512;
P. B. Fernandes, R. N. Swanson, D. J. Hardy, C. W. Hanson, L. Cohen,
R. R. Rasmussen and R. H. Chen, J. Antibiot., 1989, 42, 521;
J. P. Karwowski, M. Jackson, R. J. Theriault, R. H. Chen, G. J. Barlow
and M. L. Maus, J. Antibiot., 1989, 42, 506 and mureidomycins:
F. Isono, Y. Sakaida, S. Takahashi, T. Kinoshita and M. Inukai,
J. Antibiot., 1993, 46, 1203; F. Isono, M. Inukai, S. Takahashi,
T. Haneishi, T. Kinoshita and H. Kuwano, J. Antibiot., 1989, 42, 667;
F. Isono, T. Katayama, M. Inukai and T. Haneishi, J. Antibiot., 1989,
42, 674; K. Isono, J. Antibiot., 1988, 41, 1711; M. Inukai, F. Isono,
S. Takahashi, R. Enokita, Y. Sakaida and T. Haneishi, J. Antibiot.,
1989, 42, 662.
9 G. M. Keseru and M. Nogradi, Natural Products by Oxidative
Coupling, Biosynthesis and Synthesis, in Studies in Natural Products
Chemistry, ed. Atta-Ur-Rahman, Elsevier Science B. V., Amsterdam,
1998, vol. 20 (part F), pp. 263–322.
10 R. L. Franck and R. P. Seven, J. Am. Chem. Soc., 1949, 71, 2629.
11 For examples of ytterbium triflate used as a Lewis acid see this review
article: S. Luo, L. Zhu, A. Talukdar, G. Zhang, X. Mi, J.-P. Cheng and
P. G. Wang, Mini-Rev. Org. Chem., 2005, 2, 177 and other selected
examples: K. Manabe, D. Nobutou and S. Kobayashi, Bioorg. Med.
Chem., 2005, 13, 5154; N. Sakai, D. Aoki, T. Hamajima and
T. Konakahara, Tetrahedron Lett., 2006, 47, 1261; L.-M. Wang,
Y.-H. Wang, H. Tian, Y.-F. Yao, J.-H. Shao and B. Liu, J. Fluorine
Chem., 2006, 127, 1570.
In conclusion, we have proposed a new detailed biogenetic
hypothesis for haouamines that has led us to successfully achieve
the first steps of the first biomimetic total synthesis of haouamine
A. An advanced biomimetic precursor of that compound has been
obtained through a convenient and efficient multicomponent
reaction and the final two bonds that are needed to complete the
synthesis are currently being investigated in our laboratory. The
described biosynthetic hypothesis can serve as a useful framework
from which to develop a coherent and straightforward synthetic
plan towards haouamines (as well as close analogs such as
compounds 11, 12, 14) that competes with other synthetic
approaches. This example also contributes to demonstrate the
power of biomimetically inspired strategies in total synthesis.14
Notes and references
{ It is interesting to note that oxidative conditions under which radicals can
be generated on phenols to make phenolic couplings possible may also
generate free hydroxyl radicals responsible for the formation of L-meta-
tyrosine involved in our biogenetic proposal.
§ Compound 11: orange oil; Rf 5 0.40 (silica gel, CH2Cl2–MeOH 9 : 1); IR
(film, CHCl3): nmax 5 2937, 1585, 1260, 1030 cm21; 1H NMR (400 MHz,
CDCl3), d 1.25 (2 H, s), 3.27 (2 H, t, J 5 6.5 Hz), 3.67 (6 H, s), 3.86 (6 H, s),
5.15 (2 H, t, J 5 6.5 Hz), 6.55–7.45 (16 H, m), 8.47 (1 H, s), 8.75 (1 H, s);
13C NMR (100 MHz, CDCl3), d 29.6, 37.8, 55.1, 55.7, 63.5, 112.5, 113.5,
114.2, 116.6, 118.2, 119.6, 121.1, 122.6, 130.2, 130.8, 134.0, 136.9, 139.8,
140.5, 141.3, 160.2, 160.6; MS (ES) m/z 546 (M+), 440 (10), 426 (100);
HRMS (ES) calcd for C36H36NO4+: 546.2639, found: 546.2642.
" Compound 12: brown varnish; Rf 5 0.30 (silica gel, CH2Cl2–MeOH 85 :
15); IR (film, CH3OH): nmax 5 3061, 2926, 1589, 1029 cm21 1H NMR
;
12 L.-B. Yu, D. Chen, J. Li, J. Ramirez, P. G. Wang and S. G. Bott, J. Org.
Chem., 1997, 62, 208; B. B. Snider and B. J. Neubert, Org. Lett., 2005, 7,
2715.
(400 MHz, CD3OD), d 1.25 (2 H, s), 3.35 (2 H, t partially hidden by
CD3OD signal, J y 6.6 Hz), 4.97 (2 H, t, J 5 6.6 Hz), 6.55–7.45 (16 H, m),
8.78 (1 H, s), 8.88 (1 H, s); 13C NMR (100 MHz, CD3OD), d 30.8, 38.4,
64.4, 115.4, 115.6, 117.0, 117.2, 118.4, 119.7, 121.1, 121.3, 131.3, 131.9,
136.0, 138.5, 141.5, 141.9, 142.8, 159.3, 159.8; MS (ES) m/z 490 (M+), 398
(10), 384 (100); HRMS (ES) calcd for C32H28NO4+: 490.2013, found:
490.2015.
I Compound 14: yellow varnish; Rf 5 0.70 (silica gel, CH2Cl2–MeOH
95 : 5); IR (film, CHCl3): nmax 5 3010, 1579, 1300, 1108 cm21; 1H NMR
(400 MHz, CDCl3), d 2.52 (2 H, d, J 5 12 Hz), 2.75–3.10 (7 H, m), 3.25
(1 H, d, J 5 14.5 Hz), 3.82 (12 H, m), 6.26 (1 H, s), 6.75–7.30 (16 H, m);
13C NMR (100 MHz, CDCl3), d 30.2, 39.5, 40.1, 55.3, 58.9, 62.6, 66.0,
111.2, 111.4, 113.2, 114.4, 117.4, 117.8, 117.9, 128.5, 129.4, 129.7, 139.8,
13 K. Jakubowicz, Y.-S. Wong, A. Chiaroni, M. Be´ne´chie and
C. Marazano, J. Org. Chem., 2005, 70, 7780 and references cited therein.
14 For recent examples of biomimetic total syntheses and synthetic
approaches of complex polycyclic molecules in the alkaloid series, see
inter alia: E. Gravel, E. Poupon and R. Hocquemiller, Org. Lett., 2005,
7, 2497; C. H. Ge, S. Hourcade, A. Ferdenzi, A. Chiaroni, S. Mons,
B. Delpech and C. Marazano, Eur. J. Org. Chem., 2006, 18, 4106;
J. Sperry and C. J. Moody, Chem. Commun., 2006, 22, 2397; P. S. Baran,
B. D. Hafensteiner, N. B. Ambhaikar, C. A. Guerrero and
J. D. Gallagher, J. Am. Chem. Soc., 2006, 128, 8678.
This journal is ß The Royal Society of Chemistry 2007
Chem. Commun., 2007, 719–721 | 721