Determination of absolute configuration of Pandanus alkaloid, pandamarilactonine-A, by first asymmetric total synthesis

Article abstract of DOI:10.1016/j.tetlet.2005.06.148

The absolute configuration of pandamarilactonine-A, a pyrrolidine alkaloid, was established on the basis of total synthesis starting from l-prolinol. An insight into the mechanism of the low enantiomeric purity of natural pandamarilactonine-A is discussed

Full text of DOI:10.1016/j.tetlet.2005.06.148

Tetrahedron
Letters
Tetrahedron Letters 46 (2005) 5795–5797
Determination of absolute configuration of Pandanus alkaloid,
pandamarilactonine-A, by first asymmetric total synthesis
Hiromitsu Takayama,* Rie Sudo and Mariko Kitajima
Graduate School of Pharmaceutical Sciences, Chiba University, 1-33 Yayoi-cho, Inage-ku, Chiba 263-8522, Japan
Received 1 June 2005; revised 27 June 2005; accepted 29 June 2005
Available online 14 July 2005
Abstract—The absolute configuration of pandamarilactonine-A, a pyrrolidine alkaloid, was established on the basis of total synthe-
sis starting from L-prolinol. An insight into the mechanism of the low enantiomeric purity of natural pandamarilactonine-A is
discussed.
Ó 2005 Elsevier Ltd. All rights reserved.
Previously, we have reported the isolation of two new
pyrrolidine alkaloids, pandamarilactonine-A (1) and -B
(2), from Pandanus amaryllifolius.1 Their structures were
first characterized by spectroscopic analysis and bio-
mimetic total synthesis through a plausible biogenetic
intermediate (3),2 and the relative stereochemistry of
the vicinal chiral centers at C14 and C15 was elucidated
gave the same adducts in 80% yield, the diastereoselec-
tivity of which was 2:1. The erythro isomer (6), which
had an undesired stereochemistry, was transformed into
the threo isomer (7) by inversion of the secondary alco-
hol in it. Initially, the intramolecular Mitsunobu reac-
tion was applied to carboxylic acid (8) that was
prepared by the alkaline hydrolysis of 6. Treatment of
8 with di-tert-butyl azodicarboxylate (DTAD) and
PPh3 in THF at room temperature gave the lactone
derivative in quantitative yield, which contained threo
isomer (9) and its erythro isomer in the ratio of 7.3:1.
The unexpected minor erythro-lactone with retention
of the configuration of the alcohol was formed via an
acyloxyphosphonium ion intermediate.6 Application of
the conventional intermolecular Mitsunobu reaction to
ester (6) resulted in the recovery of the starting material.
For the inversion of the alcohol in 6, we also attempted
the oxidation–reduction sequence. Ketone derivative
(10) prepared from 6 by oxidation with Dess–Martin
periodinane was reduced with NaBH4 in MeOH at
ꢁ20 °C to afford threo (7) and erythro (6) alcohols in
the ratio of 2.6:1. The major alcohol 7 thus obtained
was treated with trifluoroacetic acid in CH2Cl2 to give
lactone (9) in 90% yield, which was identical with the
major product obtained via the intramolecular Mitsun-
obu reaction described above. Next, the isomerization
of the double bond in 9 from exo to endo position was
performed by using Et3SiH (5 mol%) and tris(triphenyl-
phosphine)rhodium chloride (10 mol%)7 in refluxing
toluene to afford a-methyl butenolide (11) in 86% yield.
by the total synthesis of racemates
1
and 2.3
Interestingly, natural pandamarilactonine-A exhibited
23
½aꢀD +35 (c 4.37, CHCl3) with 26% enantiomeric excess,
whereas pandamarilactonine-B was isolated as a race-
mate. To elucidate the absolute configuration of the ma-
jor enantiomer in pandamarilactonine-A and to examine
why pandamarilactonine-A was isolated as a compound
possessing low optical purity, we attempted the asym-
metric total synthesis of 1 and used it for resolving the
issue of concern (Fig. 1).
Initially, L-prolinol (4) was converted into aldehyde (5,
26
½aꢀD ꢁ63.1 (c 1.14, MeOH))4 in two steps, and then
the Reformatsky-type condensation with ethyl 2-(bromo-
methyl)acrylate was carried out. When Zn metal was
used, the adducts were obtained in 52% yield, which
contained erythro (more polar, 6) and threo (less polar,
7) isomers in the ratio of 4:1, the stereochemistry of
which was determined in the later stage, as described be-
low. On the other hand, indium-mediated coupling5
Keywords: Pyrrolidine alkaloid; Asymmetric total synthesis; Absolute
configuration; Optical purity.
25
1
The H and 13C NMR spectra of 11 (½aꢀD ꢁ183 (c 0.49,
CHCl3), 100% ee based on the chiral HPLC analysis)
were identical with those of Martin et al.Õs racemic
*
Corresponding author. Tel./fax: +81 43 290 2901; e-mail: htakayam@
p.chiba-u.ac.jp
0040-4039/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2005.06.148

5796
H. Takayama et al. / Tetrahedron Letters 46 (2005) 5795–5797
O
O
H
N
O
O
O
O
O
O
H
N
14
O
15
H
O
N
O
O
H
H
Pandamarilactonine-A (1) Pandamarilactonine-B (2)
Pandanamine (3)
Figure 1.
24
compound,8 the relative stereochemistry at C14 and C15
of which was established to be threo by X-ray analysis.
Careful removal of the Cbz group on nitrogen (TMSI
in CH3CN at ꢁ15 °C for 30 min) gave secondary amine
(12) (½aꢀD ꢁ49.4 (c 1.22, CHCl3))9 in quantitative yield
(Scheme 1).
The final stage of the total synthesis of 1 is the coupling
of optically active amine (12) with the C9 unit involving
a c-alkylidene butenolide moiety. Iodide (13),3 which
consisted of Z and E isomers in the ratio of 5:1, was con-
densed with freshly prepared 12 in CH3CN in the pres-
ence of Ag2CO3 at room temperature to furnish the
adducts in 66% yield. After repeated column chromato-
graphy, pure pandamarilactonine-A (1) was obtained in
48% yield. The synthetic compound with 14S and 15S
Cbz
H
H
N
H
N
i
CHO
OH
5
4
ii
1
configurations, whose spectroscopic data, including H
Cbz
and 13C NMR, UV, IR, MS, and HR–MS, were com-
Cbz
H
OH
N
pletely identical with those of the natural product,
H
OH
N
CO2Et
CO2R
23
exhibited ½aꢀD ꢁ94.0 (c 0.12, CHCl3). This finding dem-
+
H
H
onstrated that the absolute configuration of the major
enantiomer in natural pandamarilactonine-A (1) was
14R and 15R.10
7
6: R=Et
iv 8: R=H
vi
iii
vii
v
Cbz
Cbz
O
H
N
As mentioned in the introductory part, pandamarilacto-
nine-A (1) was obtained as a compound with 26%
enantiomeric excess from nature, whereas pandamaril-
actonine-B (2) was isolated as a racemate. Craik and
co-workers have claimed in their recent report11 that
pandamarilactonines isolated from acid–base treatment
are artifacts produced during the extraction of pandan-
amine (3). However, the fact that our natural pandama-
rilactonine-A was optically active, albeit its low
enantiomeric excess, led us to consider that 1 is pro-
duced enzymatically in the plants, and during the extrac-
tion or isolation process, optically active 1 might
isomerize to optically inactive 1 and 2 under acidic con-
ditions through the mechanism shown in Figure 2. To
examine this hypothetical mechanism, the behavior of
pandamarilactonine-A in acidic solution (CD3OD + 1
H
O
O
N
CO2Et
H
10
9
viii
Cbz
H
H
O
O
H
N
O
N
O
ix
H
H
11
12
O
x
O
O
O
O
13
I
1
mol% CD3CO2D) was monitored by H NMR spectro-
H
N
14
O
15
scopy. If the ring-opening/ring-closing process occurred
in this solution, the formation of both 1 and 2 having a
deuterium at the C15 position would be observed. How-
ever, the formation of 2 as well as 1 having a deuterium
at C15 was not recognized even after one month, and
only the gradual production of 3 was observed. This
indicates that pandamarilactonine-B (2) was not derived
from pandamarilactonine-A (1). Taking Craik and
co-workers experiments, that is, racemic pandamarilac-
tonines are formed from pandanamine by acid–base
treatment, into consideration, we speculate that our
H
(–)-Pandamarilactonine-A (1)
Scheme 1. Reagents and conditions: (i) Cbz–Cl, K2CO3, CH3CN, then
Swern oxidation, 91%; (ii) ethyl 2-(bromomethyl)acrylate, 2 equiv Zn,
THF–aq satd NH4Cl or 1.1 equiv indium, aq EtOH; (iii) LiOH, aq
THF, quant; (iv) DTAD, PPh3, THF, rt; (v) DMP, CH2Cl2, rt, quant;
(vi) NaBH4, MeOH, ꢁ20 °C, 86%; (vii) TFA, CH2Cl2, rt, 90%; (viii)
5 mol% Et3SiH, 10 mol% Rh(PPh3)3Cl, toluene, reflux, 86%; (ix)
TMSI, CH3CN, ꢁ15 °C, quant; (x) Ag2CO3, CH3CN, rt.

H. Takayama et al. / Tetrahedron Letters 46 (2005) 5795–5797
5797
O
O
O
O
O
O
O
O
H+
+
H+ (D+)
H
O
N
O
O
N
O
15
H
N
O
15
H
H
N
O
15
O
O
D
D
H
( )-(2)
Pandanamine (3)
(–)-(1)
( )-(1)
Figure 2. A hypothetical mechanism for isomerization of pandamarilactonine.
7. Tanaka, M.; Mitsuhashi, H.; Maruno, M.; Wakamatsu, T.
Chem. Lett. 1994, 1455–1458.
8. Martin, S. F.; Barr, K. J.; Smith, D. W.; Bur, S. K. J. Am.
Chem. Soc. 1999, 121, 6990–6997.
natural pandamarilactonine-A consists of enzymatically
formed 1 with high optical activity and a racemate
formed during the extraction/isolation process, resulting
in the compound possessing a low optical purity. By
contrast, pandamarilactonine-B is an artifact from pan-
danamine (3), as mentioned in Ref. 11.
9. In the process of the total synthesis of pandamarilacto-
nines, the Barcelona group observed the epimerization/
racemization when ethyl carbamate in the compound
similar to 9 was removed using TMSI in CHCl3 under
reflux for 5 h. (a) Busque, F.; March, P.; Figueredo, M.;
Font, J.; Sanfeliu, E. Tetrahedron Lett. 2002, 43, 5583–
5585; (b) Blanco, P.; Busque, F.; March, P.; Figueredo,
M.; Font, J.; Sanfeliu, E. Eur. J. Org. Chem. 2004, 48–53.
10. The optical purity of synthetic 1 was estimated to be 70%
ee based on the specific rotation of the natural 1. For
reference, optically active pandamarilactonine-B (2) with
14S and 15R was also prepared from 6 via the same
synthetic sequence as that for 1. A synthetic intermediate
corresponding to norpandamarilactonine-A12 showed
References and notes
1. Takayama, H.; Ichikawa, T.; Kuwajima, T.; Kitajima, M.;
Seki, H.; Aimi, N.; Nonato, M. G. J. Am. Chem. Soc.
2000, 122, 8635–8639.
2. Takayama, H.; Ichikawa, T.; Kitajima, M.; Aimi, N.;
Lopez, D.; Nonato, M. G. Tetrahedron Lett. 2001, 42,
2995–2996.
3. Takayama, H.; Ichikawa, T.; Kitajima, M.; Nonato, M.
G.; Aimi, N. Chem. Pharm. Bull. 2002, 50, 1303–1304.
4. St-Denis, Y.; Chan, T. H. J. Org. Chem. 1992, 57, 3078–
3085.
5. (a) Nair, V.; Ros, S.; Jayan, C. N.; Pillai, B. S. Tetrahedron
2004, 60, 1959–1982; (b) Chappell, M. D.; Halcomb, R. L.
Org. Lett. 2000, 2, 2003–2005; (c) Steurer, S.; Podlech, J.
Eur. J. Org. Chem. 1999, 1551–1560.
24
½aꢀD +80.2 (c 0.79, CHCl3). Except for the optical rota-
24
tion, ½aꢀD +26.9 (c 0.17, CHCl3), the spectroscopic data of
synthetic 2 were identical with those of the natural
product.1
11. Salim, A. A.; Garson, M. J.; Craik, D. J. J. Nat. Prod.
2004, 67, 54–57.
12. Takayama, H.; Ichikawa, T.; Kitajima, M.; Nonato, M.
G.; Aimi, N. J. Nat. Prod. 2001, 64, 1224–1225.
6. Ahn, C.; Deshong, P. J. Org. Chem. 2002, 67, 1754–1759.