Concise biomimetic total syntheses of both antipodes of balasubramide

Article abstract of DOI:10.1055/s-2007-986670

A two-step, protecting-group-free synthesis of the natural product balasubramide, using an Yb(OTf)₃-catalyzed intramolecular epoxide opening, is reported. Both enantiomers of the natural product are available from the antipodal forms of the sta

Full text of DOI:10.1055/s-2007-986670

LETTER
2593
Concise Biomimetic Total Syntheses of Both Antipodes of Balasubramide
A
M
B
iomimetic
S
ynthes
i
is of (+
c
)-Balasubr
a
h
mide ael B. Johansen, Andrew B. Leduc, Michael A. Kerr*
Department of Chemistry, The University of Western Ontario, London, ON, N6A 5B7, Canada
Fax +1(519)6613022; E-mail: makerr@uwo.ca
Received 8 June 2007
rived from a similar cinnamide, as evidenced by its co-iso-
lation with prebalamide (7). In this case, the indole does
not bear a 5-hydroxyl group, which would likely preclude
the formation of a C4-centered radical. Instead, the
favored biosynthetic pathway apparently involves an
intramolecular electrophilic aromatic substitution of the
epoxide. In recent years, our group has been successful in
developing methods for the functionalization of the indole
Abstract: A two-step, protecting-group-free synthesis of the
natural product balasubramide, using an Yb(OTf) -catalyzed
3
intramolecular epoxide opening, is reported. Both enantiomers of
the natural product are available from the antipodal forms of the
starting epoxycinnamic acid.
Key words: balasubramide, biomimetic synthesis, lanthanide
catalysis, chiral resolution, indole natural product
4
ring via lanthanide triflate catalysis. We felt that such
catalysis would be ideally directed towards the synthesis
of balasubramide in the biomimetic fashion shown in
Scheme 2. Moreover, a route proceeding via a homochiral
epoxide would result in an asymmetric synthesis of the
natural product in either antipodal form (providing of
course that both epoxide enantiomers are accessible). Our
efforts, culminating in the successful synthesis of both
antipodes of balasubramide, are described herein.
Recently, we reported the total synthesis of the antimalar-
ial indole natural product decursivine (1, Scheme 1). Its
1
biogenesis is likely to include the serotonin-derived
cinnamide 2 via a C4-centered (indole numbering) radical
cyclization. This hypothesis is strengthened by the co-iso-
lation of the closely related serotobenine (5), moschamin-
2
indolol (4), and moschamine (3).
O
O
O
O
N
NH
O
H
H
O
O
N
N
H
H
OH
O
O
O
HO
N
N
H
H
N
N
H
H
6: balasubramide
7: prebalamide
1: decursivine
2
Scheme 2 A proposed biomimetic retrosynthesis
HO
OMe
Though isolated a decade ago, there has been only one
HO
OMe
HO
OMe
5
(
very recent) synthesis of ent-balasubramide to date. This
synthesis was completed in eight steps, in reasonable
yield and high enantiomeric purity; however, the synthe-
sis relied on a biotransformation which was incapable of
producing the natural enantiomer.
O
H
H
O
H
N
N
N
H
H
HO
H
HO
O
H
O
HO
Our initial synthetic efforts involved simply coupling cin-
namic acid with tryptamine to form a cinnamic amide 8
N
N
N
H
H
H
(
Scheme 3); however, all attempts at epoxidation of this
3
: moschamine
4: moschaminindolol
5: serotobenine
amide to yield 7 resulted only in recovery of starting ma-
terial or decomposition products. It was therefore decided
to first epoxidize the acid residue and then couple it to
tryptamine (Scheme 4).
Scheme 1 Biosyntheses of decursivine and serotobenine
During synthetic efforts towards decursivine, we became
aware of a related natural product balasubramide (6,
Scheme 2) isolated from Clausena indica. It is also de-
To avoid problems surrounding the general instability of
epoxyacids, the product was precipitated as the potassium
salt following DMDO-mediated epoxidation of cinnamic
acid (9). While the exploratory work was conducted us-
ing racemic epoxide 10, it was always the goal to prepare
3
6
SYNLETT 2007, No. 16, pp 2593–2595
0
1
.1
0
.2
0
0
7
Advanced online publication: 12.09.2007
DOI: 10.1055/s-2007-986670; Art ID: S04207ST
optically pure epoxide to provide access to either enantio-
©
Georg Thieme Verlag Stuttgart · New York

2
594
M. B. Johansen et al.
LETTER
NH
NH
the eight-membered lactam was observed as the sole
regioisomer under these conditions, with no evidence for
the seven-membered lactam observed. That the eight-
membered ring was formed was evident based on
extensive 2D NMR analysis (HMBC, HSQC, COSY).
This regioselectivity is not surprising based on electronic
considerations of the epoxide opening.
various
O
O
epoxidation
methods
O
N
N
H
H
8
7
no reaction, or
decomposition
With this promising result it was decided that it would be
most convergent to begin the synthesis of balasubramide
with commercially available N-methyltryptamine and
thus avoid postcyclization installation of the methyl
moiety. Coupling of N-methyltryptamine and cinnamic
epoxide 10 was accomplished utilizing the same mixed
anhydride used previously. Despite our best efforts, the
resultant amide 11 could not be separated from at least one
other unidentified compound, but the mixture could once
again be subjected to microwave irradiation in the
Scheme 3 The biomimetic synthesis of balasubramide
mer of the target. Coupling of epoxycinnamic acid with
tryptamine was far more troublesome than anticipated and
was attempted under varying conditions, including the use
of DCC (which led solely to decomposition), and forma-
tion of the acid chloride, which produced only minor
amounts of product. The formation of a mixed anhydride
with pivaloyl chloride proved to be the most successful
method for the formation of the amide 7, the aforemen-
tioned prebalamide (Scheme 4). In the experiment, a sim-
ple acid workup of 10 followed by treatment of the
presence of catalytic Yb(OTf) to produce racemic bala-
3
subramide (6) in 47% yield over two steps.
With the racemic synthesis of balasubramide completed,
methods for obtaining enantiopure 10 were investigated,
for this would allow access to either isomer of bala-
subramide. A method to resolve and isolate both enantio-
mers of 10 has been previously reported utilizing both
antipodes of a-methylbenzylamine to sequentially precip-
resultant acid with Et N and pivaloyl chloride produced
3
the mixed anhydride in situ. Addition of tryptamine
directly into the reaction mixture produced the desired
prebalamide 7 with an acceptable yield of 52%, but as an
inseparable mixture contaminated with the pivaloyl amide
of tryptamine. Compound 7 could be obtained cleanly
through the use of a TFA-derived mixed anhydride, albeit
in a reduced yield of 30%.
7
itate each enantiomer. Recrystallization of the resultant
salts led to both requisite epoxides in >95% ee. The enan-
tiopure salts were then dissolved in ethanol, reprecipitated
with KOH, and taken through the previous synthetic steps
1
. NaHCO3, acetone, H2O
then Oxone, EDTA
. KOH, EtOH
(
Scheme 5). Chiral HPLC analysis demonstrated that this
approach led to the production of balasubramide (6) in
8% ee and ent-6 in 95% ee.
O
O
O
2
OH
OK
9
83% overall
9
(±)-10
O
1. HCl, Et2O
H
CO2K
2
. Me3CCOCl, Et3N, CH2Cl2
then tryptamine or N-methyltryptamine
resolution with
α-methylbenzylamine
O
Ph
H
O
(
+)-10
OK
ref. 7
O
R
R
(±)-10
Ph
H
N
Yb(OTf)3 (10 mol%),
N
O
H
CO K
CH2Cl2, microwave
80 °C), 10 min
2
(
O
(
–)-10
OH
N
N
H
O
Me
O
H
O
N
H
CO2K
as in Scheme 2
42% overall
8% ee
Ph
H
(
(
±)-12 (norbalasubramide): R = H (19% from 11)
±)-6 (balasubramide): R = Me (47% from 11)
(±)-7: R = H
(±)-11: R = Me
(+)-10
OH
N
9
H
Scheme 4 Synthesis of racemic balasubramide
found: [α]D (CHCl3, c = 0.5) = +6.9°
ref 3: [α]D (CHCl3, c = 0.5) = +7°
(
+)-6: natural balasubramide
With prebalamide 7 in hand, attempts were made to
cyclize the amide to form norbalasubramide (12). We
were pleased to observe that cyclization took place under
microwave irradiation and in the presence of catalytic
ytterbium(III)triflate producing 12 (racemic norbala-
subramide) in 37% yield. Alternate Lewis acids such as
Me
N
O
H
O
Ph
as in Scheme 2
H
CO2K
OH
(
–)-10
46% overall
N
H
Sc(OTf) were screened for this reaction, but commonly
resulted in decomposition even at room temperature.
Cyclization under microwave irradiation in the absence of
95% ee
found: [α]D (CHCl3, c = 0.5) = –6.7°
3
(–)-6: ent-balasubramide
catalyst was also unsuccessful. It was quite fortuitous that Scheme 5 Synthesis of either antipode of balasubramide
Synlett 2007, No. 16, 2593–2595 © Thieme Stuttgart · New York

LETTER
A Biomimetic Synthesis of (+)-Balasubramide
2595
In conclusion, we have successfully, and biomimetically 2.84 (s, 3 H). 13C NMR (100 MHz, CDCl
): d = 173.5, 141.0, 135.2,
3
1
7
1
32.2, 129.3, 128.7, 127.5, 127.2, 122.1, 119.4, 117.5, 110.7, 106.6,
3.8, 54.3, 46.4, 34.2, 22.8. IR (thin film): nmax = 3301, 2926, 1640,
495, 1461, 1392, 1360, 1340, 1183, 1066, 909, 734, 700. HRMS:
completed the first total synthesis of natural (+)-bala-
subramide and only the second synthesis of (–)-balasubra-
mide in two steps from epoxycinnamic acid and in
excellent overall yields of 42–46%. Most importantly, this
short synthesis of both enantiomers should allow for the
m/z calcd: 320.1525; found: 320.1530. [a] (lit.) +7 (c 0.5, CHCl ),
D
3
20
[
a]D +6.9 (c 0.5, CHCl ). The spectral and physical data for syn-
3
thetic balasubramide (6) match the literature data in all respects in-
synthesis of sufficient quantities of balasubramide or cluding optical rotation, and were further verified by 2D NMR. The
related compounds for biological screening.
enantiomer of balasubramide (ent-6) was found to have the same
spectral and physical data but with an equal and opposite optical ro-
2
0
tation [a]D –6.7 (c 0.5, CHCl3).
General Procedure for Amide Formation: Preparation of Com-
pound 7
A solution of chiral or racemic potassium salt 10 (227 mg, 1.12
Acknowledgment
We would like to thank the Natural Science and Engineering Re-
search Council of Canada (NSERC), Merck-Frosst, and Boehringer
Ingelheim Canada for their generous funding of this project. We are
grateful to Mr. Doug Hairsine for conducting MS analyses. We are
also grateful to Dr. Michael Chong of the University of Waterloo for
the use of his polarimeter. A.B.L. and M.B.J. are recipients of
NSERC scholarships (PGSD and PGSM, respectively).
mmol) in 5 mL distilled H O and ice was acidified with 3 mL of
2
1
M HCl and extracted with Et O (3 × 10 mL). The extracts were
2
dried over anhyd Na SO and filtered directly into a 100 mL round-
2
4
bottom flask. To this solution was added Et N (170 mg, 1.68 mmol)
3
and 20 mL of anhyd CH Cl to increase solubility. Pivaloyl chloride
2
2
(68.0 mg, 0.56 mmol) was added and the reaction was allowed to
stir for 2 h. After 2 h, N-methyltryptamine (97.6 mg, 0.56 mmol)
was added and the reaction was allowed to stir for a further 2 h.
Regardless of the purification procedure employed, amide 11 was
inseparable from at least one other compound. It was thus decided
to proceed through cyclization with the mixture.
References
(1) Leduc, A. B.; Kerr, M. A. Eur. J. Org. Chem. 2007, 237.
(
2) (a) Sarker, S. D.; Savchenko, T.; Whiting, P.; Pensri, S.;
Dinan, L. N. Nat. Prod. Lett. 1997, 9, 189. (b) Sato, H.;
Kawagishi, H.; Nishimura, T.; Yoneyama, S.; Yoshimoto,
Y.; Sakamura, S.; Furusaki, A.; Katsuragi, S.; Matsumoto,
T. Agric. Biol. Chem. 1985, 49, 2969.
General Procedure for Cyclization: Preparation of Com-
pound 6
A solution of amide 11 (127 mg in 5 mL CH Cl ) was treated with
2
2
–
2
Yb(OTf) (25.0 mg, 4.03·10 mmol) and irradiated in a microwave
3
for 10 min at 80 °C. Once the reaction had cooled it was pre-ab-
sorbed onto silica, and then subjected to flash column chromatogra-
phy with gradient elution beginning at 40% EtOAc in hexanes and
increasing in 5% intervals to 50% EtOAc in hexanes yielding 6 in
(3) Riemer, B.; Hofer, O.; Greger, G. Phytochemistry 1997, 45,
337.
(4) (a) England, D. B.; Kuss, T. D.; Keddy, R. G.; Kerr, M. A.
J. Org. Chem. 2001, 66, 4704. (b) Kerr, M. A.; Keddy, R. G.
Tetrahedron Lett. 1999, 5671. (c) Harrington, P. E.; Kerr,
M. A. Can. J. Chem. 1998, 76, 1256. (d) Harrington, P. E.;
Kerr, M. A. Tetrahedron Lett. 1997, 38, 5949. (e) Kerr, M.
A.; Harrington, P. E. Synlett 1996, 1047.
(5) Yang, L.; Deng, G.; Wang, D.-X.; Huang, Z.-T.; Zhu, J.-P.;
Wang, M.-X. Org. Lett. 2007, 9, 1387.
(6) Corey, P. F.; Ward, F. E. J. Org. Chem. 1986, 51, 1926.
(7) Harada, K. J. Org. Chem. 1966, 31, 1407.
4
6% over the two steps (82.7 mg, 0.25 mmol); mp 185–191 °C (de-
1
comp.); R = 0.31 (70% EtOAc in hexanes). H NMR (600 MHz,
f
CDCl ): d = 7.92 (br s, 1 H), 7.53 (dd, J = 6.8, 1.6 Hz, 1 H), 7.33–
3
7
1
.24 (m, 5 H), 7.20 (dd, J = 6.8 Hz, 2 Hz, 1 H), 7.15 (ddd, 7.8, 6.6,
.2 Hz, 1 H), 7.12 (ddd, J = 8.4, 6.8, 1.2 Hz, 1 H), 4.96 (br d, J = 6.6
Hz, 1 H), 4.37 (d, J = 6.0 Hz, 1 H), 4.34 (v br s), 3.96 (ddd, J = 15.6,
0.2, 6.6 Hz, 1 H), 3.49 (ddd, J = 16.8, 10.8, 7.8 Hz, 1 H), 3.41 (ddd,
J = 15.0, 7.2, 3.6 Hz, 1 H), 3.03 (ddd, J = 15.6, 5.4, 3.6 Hz, 1 H),
1
Synlett 2007, No. 16, 2593–2595 © Thieme Stuttgart · New York