A chemoenzymatic total synthesis of the structure assigned to the alkaloid (+)-montabuphine

Article abstract of DOI:10.1021/ol801815k

(Chemical Equation Presented) An enantioselective synthesis of the structure, 3, assigned to the alkaloid (+)-montabuphine has been achieved using the readily available metabolite 4 as starting material. A comparison of the physical and spectral data recorded on compound 3 with those reported for (+)-montabuphine suggests that they are different compounds.

Full text of DOI:10.1021/ol801815k

ORGANIC
LETTERS
2008
Vol. 10, No. 20
4693-4696
A Chemoenzymatic Total Synthesis of
the Structure Assigned to the Alkaloid
(+)-Montabuphine
Maria Matveenko, Martin G. Banwell,* and Anthony C. Willis
Research School of Chemistry, Institute of AdVanced Studies, The Australian National
UniVersity, Canberra, ACT 0200, Australia
mgb@rsc.anu.edu.au
Received September 4, 2008
ABSTRACT
An enantioselective synthesis of the structure, 3, assigned to the alkaloid (+)-montabuphine has been achieved using the readily available
metabolite 4 as starting material. A comparison of the physical and spectral data recorded on compound 3 with those reported for (+)-
montabuphine suggests that they are different compounds.
The montanine alkaloids are a small group of natural products
isolated from various Amaryllidaceae species and character-
ized by the presence of a 5,11-methanomorphanthridine
framework incorporating, in varying configurations, hydroxy
and/or methoxy groups at C2 and C3 or at C3 and C4.1 A
modest number of biological properties have been attributed
to these alkaloids including anticonvulsive, antidepressive,
anxiolytic and weak hypertensive activities.2 Representative
members of the class, which include (-)-brunsvigine (1,
Figure 1) and (-)-nangustine (2), have been the subject of
various synthetic studies that have served to confirm their
structures.3 The isolation of (+)-montabuphine, which was
assigned structure 3, from Boophane flaVa, an Amarylli-
daceae species endemic to winter rainfall areas in southern
Africa, has attracted considerable attention because this
suggests that both enantiomeric forms of the montanine
alkaloid framework occur in nature.4,5 Given the unique
Figure 1. Structures assigned to the montanine alkaloids
(-)-brunsvigine, (-)-nangustine, and (+)-montabuphine.
position (+)-montabuphine appears to hold within the class it
is surprising that no efforts to synthesize it have been reported
thus far. Accordingly, we now describe the total synthesis of
compound 3 in the illustrated enantiomeric form and report that
the physical and spectral data derived from this material do not
match those recorded for the natural product.
Our synthetic approach to compound 3 was based on related
ones that we have used recently to prepare the non-natural
enantiomeric forms of (-)-brunsvigine (1) and (-)-nangustine
(1) For reviews dealing with this class of alkaloid see: (a) Martin, S. F.
In The Alkaloids; Brossi, A., Ed.; Academic Press: San Diego, 1987; Vol.
30, p 251. (b) Hoshino, O. In The Alkaloids; Cordell, G. A., Ed.; Academic
Press: San Diego, 1998; Vol. 51, p 323. (c) Lewis, J. R. Nat. Prod. Rep.
20001757, and previous reviews in the series.
(2) See, for example: (a) Labran˜a, J.; Machocho, A. K.; Kricsfalusy,
V.; Brun, R.; Codina, C.; Viladomat, F.; Bastida, J. Phytochemistry 2002,
60, 847. (b) Schurman da Silav, A. F.; de Andrade, J. P.; Bevilaqua,
L. R. M.; da Souza, M. M.; Izquierdo, I.; Henriques, A. T.; Zuanazzi, J. A. S.
Pharmacol., Biochem. BehaV. 2006, 85, 148.
10.1021/ol801815k CCC: $40.75
Published on Web 09/25/2008
2008 American Chemical Society

(2), namely, ent-13k and ent-2.3l The starting material employed
in these cases, and in the present work, was the enantiopure
cis-1,2-dihydrocatechol 4 (Figure 2) which can be obtained in
large quantity via the whole-cell mediated biotransformation
of chlorobenzene.6 The synthetic sequence involves three critical
transformations, an Overman rearrangement7 to introduce the
nitrogen, a novel radical addition/elimination reaction3k,l,8 to
establishtheD-ringoftarget3andalate-stagePictet-Spengler
reaction9 to introduce the C6-methylene unit and thereby
form the C-ring.
hydride (DIBAl-H), engaged in smooth reductive cleavage to
the bis-ether 9 (68%).10 The free hydroxy group within this
last compound was protected, under standard conditions, as the
corresponding MOM-ether 10 and this was then treated with
2,3-dichloro-5,6-dicyano-1,4-benzoquinone (DDQ) so as to
cleave the PMB ether moiety and thus giving the allylic alcohol
11 (91% from 9). In anticipation of the foreshadowed Overman
rearrangement reaction, compound 11 was treated with trichlo-
roacetonitrile in the presence of sodium hydride and so affording
the pivotal but rather unstable trichloroacetimidate 12.
Exposure of an o-dichlorobenzene solution of compound 12
containing potassium carbonate to microwave irradiation (Scheme
2) afforded the anticipated rearrangement product 13 (88% from
Scheme 2
.
Synthesis of the Substrate for the Radical Addition/
Elimination Reaction
Figure 2. Starting material used in the synthesis of compound 3.
The route used in preparing the substrate required for the
Overman rearrangement is shown in Scheme 1 and started with
the known3l and readily generated PMP-acetal derivative, 5, of
cis-1,2-dihydrocatechol 4. Epoxidation of acetal 5 with
m-chloroperbenzoic acid (m-CPBA) proceeded in a regio- and
stereo-selective fashion to give the previously reported3l oxirane
6 (75% yield from 4) and this underwent selective reductive
cleavage on exposure to lithium aluminum hydride to give
alcohol 7 in 92% yield. O-Methylation of the last compound
with methyl iodide in the presence of sodium hydride afforded
acetal/ether 8 (95%) that, on exposure to di-isobutyl aluminum
Scheme 1. Substrate Synthesis for the Overman Rearrangement
11), the acetamide residue of which was immediately subjected
to reductive cleavage using DIBAl-H and thus giving the
corresponding primary amine 14 in 77% yield. Reductive
amination of this last compound with p-methoxybenzaldehyde
(p-MBA) in the presence of sodium borohydride then afforded
the expected secondary amine 15 (92%) which could be coupled
(3) (a) Overman, L. E.; Shim, J. J. Org. Chem. 1991, 56, 5005. (b)
Ishizaki, M.; Hoshino, O.; Iitaka, Y. J. Org. Chem. 1992, 57, 7285. (c)
Ishizaki, M.; Kurihara, K.-I.; Tanazawa, E.; Hoshino, O. J. Chem. Soc.,
Perkin Trans. 1 1993, 101. (d) Overman, L. E.; Shim, J. J. Org. Chem.
1993, 58, 4662. (e) Jin, J.; Weinreb, S. M. J. Am. Chem. Soc. 1997, 119,
5773. (f) Pearson, W. H.; Lian, B. W. Angew. Chem., Int. Ed. 1998, 37,
1724. (g) Ikeda, M.; Hamada, M.; Yamashita, T.; Matsui, K.; Sato, T.;
Ishibashi, H. J. Chem. Soc., Perkin Trans. 1 1999, 1949. (h) Sha, C.-K.;
Hong, A.-W.; Huang, C.-M. Org. Lett. 2001, 3, 2177. (i) Banwell, M. G.;
Edwards, A. J.; Jolliffe, K. A.; Kemmler, M. J. Chem. Soc., Perkin Trans.
1 2001, 1345. (j) Pandey, G.; Banerjee, P.; Kumar, R.; Puranik, V. G. Org.
Lett. 2005, 7, 3713. (k) Banwell, M. G.; Kokas, O. J.; Willis, A. C. Org.
Lett. 2007, 9, 3503. (l) Kokas, O. J.; Banwell, M. G.; Willis, A. C.
Tetrahedron 2008, 64, 6444.
4694
Org. Lett., Vol. 10, No. 20, 2008

with the racemic modification of the previously reported3g acid
16 using 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide (EDCI)/
1-hydroxybenzotriazole (HOBt). In this manner the substrate
required for the foreshadowed radical addition/elimination
reaction,8 namely amide 17, was obtained in 77% yield and as
a ca. 1:1 mixture of diastereoisomers.
Table 1. Comparison of the 13C NMR Data for
Naturally-Occurring (+)-Montabuphine, Synthetically-Derived 3,
and Other Montanine Alkaloid-Type Structures
montabuphine
(δC)a
3 (δC)b 23 (δC)b 25 (δC)c 26 (δC)c 27 (δC)d
The final stages of the synthesis of target compound 3 are
shown in Scheme 3 and involve the subjection of compound
150.8
146.9
146.3
130.9
122.6
117.6
107.6
106.7
100.8
77.0
153.6
146.7
145.9
132.1
124.5
116.8
107.7
106.7
100.7
77.6
154.4
146.7
145.9
132.2
124.6
114.7
107.4
106.8
100.7
81.3
156.1
146.8
145.9
132.1
125.0
111.9
107.8
106.9
100.8
77.7
154.3
146.7
145.9
132.5
124.8
112.9
107.3
106.8
100.7
79.8
154.0
146.7
145.9
131.5
124.3
116.2
107.5
106.7
100.7
75.2
Scheme 3. Completion of the Synthesis of Compound 3
67.8
68.2
67.3
65.6
69.2
74.6
60.0
61.2
61.0
61.2
60.9
63.0
58.7
58.2
58.6
58.1
58.7
61.0
57.4
57.4
57.0
56.7
57.6
-
55.1
55.7
55.6
55.7
55.4
55.5
44.8
45.5
45.6
46.0
45.6
45.1
31.6
32.8
28.8
36.0
32.7
37.8
a Data from ref 4 and recorded in CDCl3 at 50 MHz. b Data arising
from work reported in this paper and recorded in CDCl3 at 75 MHz. c Data
from ref 3e and recorded in CDCl3 at 125 MHz. d Data from ref 3d and
recorded in CDCl3 at 125 MHz.
of a chlorine radical to reinstate, in a completely regioselective
fashion, the cyclohexenyl double-bond and so deliver the
observed product 18. The origins of the pleasing diastereose-
lectivity observed in the cyclization process remain unclear at
the present time. As a prelude to carrying out the Pictet-Spengler
reaction, the lactam carbonyl within compound 18 was removed
using aluminum hydride generated in situ from AlCl3 and
LiAlH4. The PMB residue associated with the resulting pyrro-
lidine 19 (89%) was cleaved using triphosgene11 and the
carbamoyl chloride 20 so-formed was subjected to acid-
catalyzed hydrolysis and thus providing the amino-alcohol 21
in 88% yield from precursor 19. Treatment of compound 21
with a mixture of paraformaldehyde and formic acid at 80 °C
for 16 h effected the desired Pictet-Spengler reaction but this
was accompanied by formylation of the free hydroxyl group
within the substrate and such that the formate ester 22 was
obtained in 97% yield. Hydrolysis of the latter compound was
readily achieved with potassium carbonate in methanol and thus
delivering the epimer, 23, of target compound 3 in 83% yield.
The structure of compound 23 was secured through a single-
(6) Compound 4 can be obtained from the Aldrich Chemical Co.
(Catalogue Number 489492) or from Questor, Queen’s University of Belfast,
Northern Ireland (http://questor.qub.ac.uk/newsite/contact.htm). For reviews
on methods for generating cis-1,2-dihydrocatechols by microbial dihy-
droxylation of the corresponding aromatics, as well as the synthetic
applications of these metabolites, see: (a) Hudlicky, T.; Gonzalez, D.;
Gibson, D. T. Aldrichim. Acta 1999, 32, 35. (b) Banwell, M. G.; Edwards,
A. J.; Harfoot, G. J.; Jolliffe, K. A.; McLeod, M. D.; McRae, K. J.; Stewart,
S. G.; Vo¨gtle, M. Pure Appl. Chem. 2003, 75, 223. (c) Johnson, R. A. Org.
React. 2004, 63, 117.
17 to reaction with a combination of hexa-n-butylditin and tri-
n-butyltin hydride. Under such conditions the pivotal radical
addition/elimination process takes place and such that lactam
18 is obtained, as a single diastereoisomer, in 95% yield. This
reaction is presumed to involve initial homolytic cleavage of
the thiophenyl residue then cyclization, via a highly diastereo-
selective 5-exo-trig process, of the resulting benzylic radical
onto the proximate terminus of the chloroalkene. The R-chlo-
rinated cyclohexyl radical so-formed then collapses with ejection
(7) (a) Overman, L. E. Acc. Chem. Res. 1980, 13, 218. (b) Overman,
L. E.; Carpenter, N. E. Org. React. 2005661, and references cited therein.
(8) For related examples of this type of radical reaction, see: (a) Stanis-
lawski, P. C.; Willis, A. C.; Banwell, M. G. Org. Lett. 2006, 8, 2143. (b)
Stanislawski, P. C.; Willis, A. C.; Banwell, M. G. Chem. Asian J. 2007, 2,
1127.
(9) For a very useful review of the Pictet-Spengler reaction, see: Cox,
E. D.; Cook, J. M. Chem. ReV. 1995, 95, 1797.
(4) Viladomat, F.; Bastida, J.; Codina, C.; Campbell, W. E.; Mathee, S.
Phytochemistry 1995, 40, 307.
(10) The regioselectivity observed in the conversion 8 f 9 has been
rationlized elsewhere, see: Matveenko, M.; Banwell, M. G.; Willis, A. C.
Tetrahedron 2008, 64, 4817.
(5) In assigning structure 3 to (+)-montabuphine, Codina et al.4 did not
explicitly define the stereochemistry at C4a but their commentary and the
subsequent interpretations of others means that the illustrated R-configuration
at this center is now widely assumed.
(11) Banwell, M. G.; Coster, M. J.; Harvey, M. J.; Moraes, J. J. Org.
Chem. 2003, 68, 613.
Org. Lett., Vol. 10, No. 20, 2008
4695

1
Table 2. Comparison of the H NMR Data for Naturally-Occurring Montabuphine and Synthetically-Derived 3 and 23
montabuphine (δH)a
3 (δH)b
23 (δH)b
6.54 (s, 1H)
6.54 (s, 1H)
6.45 (s, 1H)
6.54 (s, 1H)
6.45 (s, 1H)
6.46 (s, 1H)
5.88 (d, J ) 1.5 Hz, 1H)
5.86 (d, J ) 1.5 Hz, 1H)
5.88 (d, J ) 1.0 Hz, 1H)
5.85 (d, J ) 1.0 Hz, 1H)
5.50 (m, 1H)
5.89 (d, J ) 1.5 Hz, 1H)
5.86 (d, J ) 1.5 Hz, 1H)
5.53 (broad s, 1H)
5.53 (dd, J ) 2.5 and 2.0 Hz, 1H)
4.38 (d, J ) 16.5 Hz, 1H)
4.18 (ddd, J ) 5.0, 3.5 and 2.5 Hz, 1H)
3.87 (d, J ) 16.5 Hz, 1H)
3.70 (ddd, J ) 5.0, 4.5 and 1.5 Hz, 1H)
3.54 (broad, J ) 13.0 Hz, 1H)
3.39 (s, 3H)
4.30 (d, J ) 16.5 Hz, 1H)
4.20 (broad s, 1H)
4.34 (d, J ) 16.5 Hz, 1H)
4.07 (broad s, 1H)
3.79 (d, J ) 16.5 Hz, 1H)
3.72 (m, 1H)
3.81 (d, J ) 16.5 Hz, 1H)
3.54 (m, 1H)
3.43 (broad m, 1H)
3.38 (s, 3H)
3.41 (s, 3H)
3.34 (m, 1H)
3.30 (d, J ) 2.0 Hz, 1H)
3.20 (broad d, J ) 1.5 Hz, 1H)
3.02 (d, J ) 11.0 Hz, 1H)
3.00 (dd, J ) 11.0 and 2.0 Hz, 1H)
2.57 (td, J ) 13.0 and 4.5 Hz, 1H)
-
3.27 (broad s, 1H)
3.11 (dd, J ) 11.0 and 2.0 Hz, 1H)
3.07 (d, J ) 11.0 Hz, 1H)
2.70 (ddd, J ) 13.0, 4.5 and 4.5 Hz, 1H)
-
3.07 (m, 1H)
3.04 (d, J ) 11.5 Hz, 1H)
2.28 (m, 1H)
1.72 (broad s, 1H, OH)
1.49 (td, J ) 13.0 and 3.5 Hz, 1H)
1.58 (ddd, J ) 13.0, 13.0 and 1.5 Hz, 1H)
a Data from ref 4 and recorded in CDCl3 at 500 MHz. b Data arising from work reported in this paper and recorded in CDCl3 at 500 MHz.
1.48 (td, J ) 13.0 and 1.5 Hz, 1H)
crystal X-ray analysis. Initial attempts to effect the conversion
of allylic alcohol 23 into its epimer involved subjecting the
former compound to a Mitsunobu reaction with R-chloroacetic
acid as nucleophile.12 However, no useful outcomes were
obtained under such conditions. Accordingly, compound 23 was
oxidized to the corresponding enone 24 (91%) using the
Dess-Martin periodinane (DMP)13 and this was then reduced
to target 3 (84% from 23) using the Luche reagent.14
The physical and spectral data derived from compound 3
were in full accord with the assigned structure15 but did not
match those reported4 for (+)-montabuphine. Thus, the specific
rotation of the synthetically-derived material is +120 (c 0.10,
Figure 3
.
Structures of compounds 25-27.
ethanol) whereas that recorded for the title alkaloid is +157 (c
0.106, ethanol). Furthermore, compound 3 was obtained as a
microcrystalline solid melting between 62 and 66 °C while (+)-
montabuphine is reported4 to have a melting range of 162 to
164 °C. While the EI mass spectra and the infrared spectra of
the two compounds compare reasonably well, the corresponding
13C NMR spectra (Table 1) do not. Most obviously, the lowest
field of the signals observed in the spectrum of the natural
product appears at δ 150.8 while in the spectrum of syntheti-
cally-derived 3 the equivalent signal appears at δ 153.6, in
keeping with the chemical shifts observed for the analogous
carbon in the related compounds 23, 25 [(-)-coccinine], 26
[(-)-montanine], and 27 (Figure 3).
1
A comparison of the H NMR spectral data recorded on
compounds 3 and 23 with those reported4 for (+)-montabuphine
is presented in Table 2. Once again, there are discrepancies
between the data sets for the natural product and the syntheti-
cally-derived material. On this basis, and given the variations
noted above, we conclude that structure 3 has been incorrectly
assigned to the alkaloid (+)-montabuphine. Work is now
underway in our laboratories to try and establish the true
structure of this natural product.
Acknowledgment. We thank the Institute of Advanced
Studies and the Australian Research Council for generous
financial support. Robert J. Dancer (H. Lundbeck A/S,
Denmark) is thanked for helpful discussions.
Supporting Information Available: Full experimental
procedures; 1H and/or 13C NMR spectra of compounds 3, 7-15,
and 18-24; single-crystal X-ray data and atomic displacement
ellipsoid plots for compound 23 and the oxalate salt of 3 (CCDC
numbers 697100 and 699621, respectively). This material is
available free of charge via the Internet at http://pubs.acs.org.
(12) Sa¨ıah, M.; Bessodes, M.; Antonakis, K. Tetrahedron Lett. 1992,
33, 4317.
(13) (a) Dess, D. B.; Martin, J. C. J. Org. Chem. 1983, 48, 4155. (b)
Dess, D. B.; Martin, J. C. J. Am. Chem. Soc. 1991, 113, 7277. A very
well-defined method for preparing this useful reagent has been reported:
(c) Boeckman, R. K., Jr.; Shao, P.; Mullins, J. J. Org. Synth. 1999, 77,
141.
(14) (a) Luche, J.-L. J. Am. Chem. Soc. 1978, 100, 2226. (b) Luche,
J.-L.; Rodriguez-Hahn, L.; Crabbe´, P. J. Chem. Soc., Chem. Commun. 1978,
601. For a useful review of lanthanide reagents in organic synthesis, see:
(c) Molander, G. A. Chem. ReV. 1992, 92, 29.
(15) The structure of synthetically-derived 3 was confirmed by a single
crystal X-ray analysis of its oxalate salt.
OL801815K
4696
Org. Lett., Vol. 10, No. 20, 2008