A Pd-mediated approach to the synthesis of an unusual β- hydroxytryptophan amino acid constituent of cyclomarin A

Article abstract of DOI:10.1055/s-2006-941560

A synthetic approach based on a palladium-mediated vinylation of indole derivatives was established for the preparation of 5, a key intermediate in the synthesis of N-(tert-butoxycarbonyl)-L-1H-[(R)-1,1-dimethyl-2,3-epoxypropyl]- β-hydroxytryptophan ethyl ester (6). Unsatisfying yields were obtained using N-alkyl-3-haloderivatives. The best method proved to be the oxidative coupling on 3-unsubstituted indole derivative. Georg Thieme Verlag Stuttgart.

Full text of DOI:10.1055/s-2006-941560

LETTER
1319
A Pd-Mediated Approach to the Synthesis of an Unusual b-Hydroxy-
tryptophan Amino Acid Constituent of Cyclomarin A
S
ynthesis of an U
n
i
usual
b
o
-Hydroxytry
r
ptophan
g
A
mino Acid
i
Consti
o
tuent of
C
yclomar
D
in
A
ella Sala, Irene Izzo, Aldo Spinella*
Dipartimento di Chimica, Università di Salerno, Via S. Allende, 84081 Baronissi (Salerno), Italy
Fax +39(089)965296; E-mail: spinella@unisa.it
Received 14 March 2006
This work is dedicated to the memory of Prof. G. Sodano.
The interesting activities displayed by this peptide have
stimulated synthetic studies4–8 and recently two approach-
es for the synthesis of 2 have been reported, both starting
from indoline (Scheme 1). In the approach of Sugijama et
Abstract: A synthetic approach based on a palladium-mediated
vinylation of indole derivatives was established for the preparation
of 5, a key intermediate in the synthesis of N-(tert-butoxycarbonyl)-
L-1H-[(R)-1,1-dimethyl-2,3-epoxypropyl]-b-hydroxytryptophan
ethyl ester (6). Unsatisfying yields were obtained using N-alkyl-3- al.,4 key steps of the synthesis are a Horner–Wadsworth–
haloderivatives. The best method proved to be the oxidative cou-
pling on 3-unsubstituted indole derivative.
Emmons (HWE) reaction and Sharpless asymmetric
aminohydroxylation (AAH). More recently, Hansen et
al.8 described another strategy based on a diastereoselec-
Key words: indoles, palladium, Heck reaction, marine natural
product, Stille reaction
tive addition of an indole Grignard 7 to a chiral serine
aldehyde equivalent 8.
During our synthetic studies on N-isoprenylindole
In 1999 Fenical and co-workers reported the isolation,
structural elucidation and potent antiinflammatory activi-
derivatives9 we have planned the preparation of 2 using
indole, instead of indoline, as starting material. To this
purpose we have systematically evaluated the application
ty of cyclomarin A (1, Figure 1), a compound produced by
a marine bacterium (Streptomyces sp.).1 Cyclomarin A is
of Pd-mediated coupling reactions to the construction of
a cyclic heptapeptide which also shows antiviral2 and
the required 3-functionalized indole and in this communi-
antimycobacterial3 activity and is currently under investi-
gation for therapeutic application. This peptide incorpo-
rates three common and four unusual amino acids. One of
cation we describe our effort to develop this alternative
synthetic strategy to 5.
Bromo compound 9, for the envisaged palladium-cata-
these new amino acids 2 is a tryptophan derivative con-
lyzed coupling, was generated from 10 whose preparation
taining a reverse prenylated moiety.
followed a described strategy starting from indole.10
Sharpless asymmetric dihydroxylation, using chiral
ligand (DHQD)2PYR, furnished diol 11 with a satisfying
enantiomeric excess (76% ee). Finally, protection of diol
O
11 as acetonide completed the preparation of 9
(Scheme 2).
OMe
O
O
N
H
HN
O
N
With the requisite bromo compound in hand we were able
to prepare compound 12 using Heck coupling. However,
the coupling reaction of 9 and ethyl acrylate proved to be
somewhat difficult when typical procedures were con-
sidered. Due to the scarce reactivity of indole substrate, a
large amount of Pd–phosphine complex was needed in or-
der to obtain satisfying yields. In fact, increasing amounts,
starting from 0.02 equivalent, were used and only using
0.3 equivalent of Pd–phosphine complex we succeeded to
obtain reasonable yields (Scheme 3). The reaction afford-
ed the expected coupling adduct together with large
amounts of the debrominated indole derivative 13 as side
product.
Me
NH
O
NH
Me
O
N
HN
OH
O
OH
N
O
cyclomarin A (1)
It is well known that N-sulfonyl-3-bromoindole and N-
acyl-3-bromoindole derivatives are in general poor sub-
strates in Heck coupling, probably due to the high electron
density of the halogenated carbon leading to a slower
oxidative addition to Pd(0)11,12 and this can explain the
disappointing yields obtained with the more electron-rich
N-alkyl-3-bromo derivatives. In view of this fact, we
Figure 1
SYNLETT 2006, No. 9, pp 1319–1322
Advanced online publication: 26.04.2006
DOI: 10.1055/s-2006-941560; Art ID: G09206ST
© Georg Thieme Verlag Stuttgart · New York
0
1.
0
6.
2
0
0
6

1320
G. Della Sala et al.
LETTER
HO
COOR
NHR'
COOEt
CHO
AAH
Sharpless
HWE
N
N
N
ref. 4
O
O
O
6 R = Et; R' = Cbz
2 R = H; R' = H
4
5
N
H
MgBr
CHO
O
O
3
ZHN
O
N
ref. 8
O
8
7
Scheme 1 Reported approaches to 2 starting from indoline.
Br
COOEt
+
Br
Br
COOEt
a
b
N
N
N
9
N
N
O
O
O
O
O
OH
O
OH
11
10
9
12
13
Scheme 2 Reagents and conditions: a) OsO4, K3Fe(CN)6,
(DHQD)2PYR, K2CO3, H2O–t-BuOH 1:1, 16 h, 0 °C, 76% ee; b)
Me2C(OMe)2, p-TsOH, r.t., 1.5 h, 85%, two steps.
Pd(OAc)2, P(o-tol)3, TEA,
80 °C, 24 h, DMF
56%
46%
41%
Pd2(dba)3, P(o-tol)3, TEA,
80 °C, 18 h, CH3CN
18%
(+ 32% of 9)
thought that the employment of the more reactive iodo de-
rivative could have reduced the amount of 13 and improve
yield of 12. No data in literature were found on Heck cou-
pling of N-alkyl-3-haloindoles, so we decided to compare
the reactivity of the model compounds 14 and 15
(Table 1).
Scheme 3
time, better yield of 16, minor amount of dehalogenated
product 17). Under Stille coupling conditions,13,14 the
yields were lower.
However, these results encouraged us to use iodoindole
derivative as coupling reagent. The trend in reactivity of
haloindoles was confirmed because 3-iodoindole 18,
obtained as depicted in Scheme 4, reacted in the Heck
Both Heck and Stille coupling reactions were assayed. 3-
Iodo-N-methylindole (15), although not a particularly
efficient substrate, was more reactive than 3-bromo-N-
methylindole (14) in the Heck reaction (shorter reaction
Table 1 Heck and Stille Coupling on N-Methyl-3-haloindoles 14 and 15
COOEt
+
X
R
COOEt
N
N
N
Pd cat.
Me
Me
Me
17
16
14: X = Br
15: X = I
X
Br
I
R
Conditions
Yield of 16 (%)
Yield of 17 (%)
1
2
3
4
5
H
Pd(OAc)2 (0.2 equiv), P(o-tol)3, Et3N, 100 °C, 24 h
Pd(OAc)2 (0.1 equiv), P(o-tol)3, Et3N, 100 °C, 2 h
Pd(PPh3)4 (0.1 equiv), LiCl, 65 °C, 18 h
36
48
41
35
–
41
H
19
I
Bu3Sn
Bu3Sn
Bu3Sn
traces
traces
–
I
Pd2(dba)3 (0.2 equiv), PPh3, CuI, LiCl, 100 °C, 4 h
PdCl2(PPh3)2 (0.1 equiv), Et4NCl, 80 °C, 44 h
Br
Synlett 2006, No. 9, 1319–1322 © Thieme Stuttgart · New York

LETTER
Synthesis of an Unusual b-Hydroxytryptophan Amino Acid Constituent of Cyclomarin A
1321
coupling with ethyl acrylate in shorter time with higher
yield (68% yield) and minor amount of dehalogenated in-
dole 13 (23%, Table 2, entry 1). As is the case of N-methyl
derivative, Stille coupling protocols (Table 2, entries 2, 3)
showed low efficiency.
O
O
N
I
Br
N
19
b
O
a
N
N
N
O
O
O
O
O
O
O
Figure 2
9
13
18
The formal synthesis was completed by deprotection of
acetonide 12 by acetic acid and transformation of the re-
sulting diol into epoxide (5) by the internal displacement
of the corresponding monotosylate (Scheme 5).
Scheme 4 Reagents and conditions: a) 1. t-BuLi, Et2O, –100 °C;
2. H2O, 92%; b) I2, KOH, DMF, r.t., 77%.
Finally, we decided to investigate the use of the Pd-cata-
lyzed oxidative coupling between indole 13 and ethyl
acrylate as an alternative protocol, to avoid the use of
halogenated substrates. An oxidative palladium(II)-medi-
ated process in the coupling of arenes with olefins was
first described by Fujiwara et al. in 1967.15 In this reaction
the coupling of arenes with olefins is promoted by the
cleavage of aromatic C–H bonds in the presence of Pd
compounds.16 The application of this methodology to in-
dole was reported in 199917 and two recent papers de-
scribed conditions for the C-2 and C-3 alkenylation of
COOEt
COOEt
a, b
N
N
O
O
O
12
5
Scheme 5 Reagents and conditions: a) AcOH 90%, THF, r.t., 7 d;
b) 1) TsCl, Et3N, CH2Cl2, 0 °C, 15 h; 2) K2CO3, EtOH, r.t., 48 h, 60%,
indole.18,19 These recent results prompted us to apply this three steps.
procedure to our substrate. Compound 13 was subjected
to coupling under Pd(OAc)2 catalysis at 80 °C, employing
In conclusion, synthesis of compound 5 was accom-
Cu(OAc)2 as the oxidant (Table 2, entry 4).20 Indole-
acrylic ester (12)21 was obtained with a good yield (70%).
This reaction afforded also considerable amounts of self-
coupling product 1922 (23%, Figure 2).
plished using a strategy centred on a Pd coupling reaction.
In our study we have shown that the best method for C-3
palladium-mediated vinylation of the indole moiety is the
oxidative Heck coupling. Since this advanced intermedi-
ate has been converted into 6,4 this synthesis constitutes a
formal synthesis of the unusual 3-hydroxytryptophan
amino acid 2.
A remarkable yield improvement to 86%, however, was
assessed with PdCl2 as catalyst at 40 °C (Table 2, entry
5).23 No traces of 19 were observed when employing
PdCl2.
Table 2 Coupling Reaction on Indole Derivatives 13 and 18
COOEt
X
R
COOEt
Pd cat.
N
N
O
O
O
O
13: X = H
18: X = I
12
X
I
R
Conditions
Yield of 12 (%)
Yield of 13 (%)
1
2
3
4
5
H
Pd(OAc)2 (0.2 equiv), P(o-tol)3, TEA, 80 °C, 5 h
Pd(PPh3)4 (0.1 equiv), LiCl, 65 °C, 5 h
68
36
39
70
86
23
2
I
Bu3Sn
Bu3Sn
H
I
Pd2(dba)3 (0.2 equiv), PPh3, CuI, LiCl, 100 °C, 6 h
Pd(OAc)2 (0.4 equiv), Cu(OAc)2, 80 °C, 5 h
PdCl2 (0.3 equiv), Cu(OAc)2, 40 °C, 5 h
15
–
H
H
H
–
Synlett 2006, No. 9, 1319–1322 © Thieme Stuttgart · New York

1322
G. Della Sala et al.
LETTER
(20) Procedure for Pd(OAc)2-Catalyzed Oxidative Heck
Acknowledgment
Coupling (Table 2, Entry 4).
The authors thank Miss Antonella D’Acunto for the help given in
the experimental work and Prof. Ricci (Universitá di Bologna) for
the useful suggestions on the oxidative palladium(II)-mediated pro-
cedure using PdCl2. Financial support from Università di Salerno is
gratefully acknowledged.
A mixture of indole 13 (90 mg, 0.35 mmol), Pd(OAc)2 (36
mg, 0.16 mmol), Cu(OAc)2 (0.318 g, 1.75 mmol) and ethyl
acrylate (0.11 mL, 1.0 mmol) in dry DMF–DMSO 9:1 (10
mL) was deoxygenated and heated at 80 °C in a capped
Schlenk tube. After 24 h, the reaction vessel was cooled to
r.t. and CH2Cl2 (20 mL) was added. The organic phase was
washed with H2O (3 × 30 mL) and the resulting aqueous
phases were again extracted with CH2Cl2 (20 mL). The
combined organic phases were dried over MgSO4, filtered
and concentrated in vacuo. Purification by flash
References and Notes
(1) (a) Renner, M. K.; Shen, Y.-C.; Cheng, X.-C.; Jensen, P. R.;
Frankmoelle, W.; Kauffmann, C. A.; Fenical, W.;
Lobkovsky, E.; Clardy, J. J. Am. Chem. Soc. 1999, 121,
11273. (b) Fenical, W. H.; Jacobs, R. S.; Jensen, P. R. U.S.
Patent 5,444,043, 1995. (c) Fenical, W. H.; Jacobs, R. S.;
Jensen, P. R. U.S. Patent 5,593,960, 1997.
chromatography (gradient elution with PE–Et2O mixtures
from 100:0 to 1:1) afforded product 12 (87 mg, 70% yield)
as a yellow oil and an additional product 19 (21 mg, 23%
yield).
(21) Compound 12: [a]D22 +12.3 (c 1.8, CHCl3). 1H NMR (400
MHz, CDCl3): d = 7.92 (1 H, m, indole H-4 or H-7), 7.90 (1
H, d, J = 16.0 Hz, b-acrylic CH), 7.70 (1 H, m, indole H-4 or
H-7), 7.62 (1 H, s, indole H-2), 7.20–7.24 (2 H overlapped,
m, indole H-5, H-6), 6.41 (1 H, d, J = 16.0 Hz, a-acrylic
CH), 4.84 (1 H, dd, J = 7.0, 6.1 Hz, acetonide CH), 4.27 (2
H, q, J = 7.1 Hz, acrylate CH2), 3.86 (1 H, dd, J = 8.9, 7.0
Hz, acetonide CH2), 3.60 (1 H, dd, J = 8.9, 6.1 Hz, acetonide
CH2), 1.79 (3 H, s, CH3), 1.72 (3 H, s, CH3), 1.41 (3 H, m,
acetonide CH3), 1.35 (3 H, t, J = 7.1 Hz, acrylate CH3), 1.34
(3 H, m, acetonide CH3). 13C NMR (100 MHz, CDCl3): d =
168.3, 138.0, 136.6, 130.8, 127.7, 122.3, 121.1, 120.7,
114.2, 112.7, 111.6, 110.3, 79.1, 65.3, 61.0, 60.0, 25.9, 24.7,
24.2, 23.5, 14.4. MS (EI, 70 eV, 250 °C): m/z = 357 [M+],
256. Anal. Calcd for C21H27NO4 (%): C, 70.56; H, 7.61; N,
3.92. Found: C, 70.27; H, 7.43; N, 3.92.
(2) Pazoles, C. J.; Siegel, S. A. U.S. Patent 5,759,995, 1998.
(3) Waters, B.; Saxena, G.; Wanggui, Y.; Kau, D.; Wrigley, S.;
Stokes, R.; Davies, J. J. Antibiot. 2002, 55, 407.
(4) Sugiyama, H.; Shioiri, T.; Yokokawa, F. Tetrahedron Lett.
2002, 43, 3489.
(5) Tarver, J. E. Jr.; Joullié, M. M. J. Org. Chem. 2004, 69, 815.
(6) Tarver, J. E. Jr.; Terranova, K. M.; Joullié, M. M.
Tetrahedron 2004, 60, 10277.
(7) Hajra, S.; Karmakar, A. Tetrahedron Lett. 2004, 45, 3185.
(8) Hansen, D. B.; Lewis, A. S.; Gavalas, S. J.; Joullié, M. M.
Tetrahedron: Asymmetry 2006, 17, 15.
(9) Della Sala, G.; Capozzo, D.; Izzo, I.; Giordano, A.;
Iommazzo, A.; Spinella, A. Tetrahedron Lett. 2002, 43,
8839.
(10) Pirrung, M. C.; Li, Z.; Park, K.; Zhu, J. J. Org. Chem. 2002,
67, 7919.
(11) Franck, W. C.; Kim, Y. C.; Heck, R. F. J. Org. Chem. 1978,
43, 2947.
(12) Harrington, P. J.; Hegedus, L. S. J. Org. Chem. 1984, 49,
2657.
(13) (a) Ahaidar, A.; Fernández, D.; Danelón, G.; Cuevas, C.;
Manzanares, I.; Albericio, F.; Joule, J. A.; Álvarez, M. J.
Org. Chem. 2003, 68, 10020. (b) Ahaidar, A.; Fernández,
D.; Pérez, O.; Danelón, G.; Cuevas, C.; Manzanares, I.;
Albericio, F.; Joule, J. A.; Álvarez, M. Tetrahedron Lett.
2003, 44, 6191. (c) Álvarez, M.; Fernández, D.; Joule, J. A.
Tetrahedron Lett. 2001, 42, 315.
(14) Sakamoto, T.; Kondo, Y.; Yasuhara, A.; Yamanaka, H.
Tetrahedron 1991, 47, 1877.
(15) Moritani, I.; Fujiwara, Y. Tetrahedron Lett. 1967, 12, 1119.
(16) Jia, C.; Piao, D.; Oyamada, J.; Lu, W.; Kitamura, T.;
Fujiwara, Y. Science 1992, 287, 1992.
(17) Jia, C.; Lu, W.; Kitamura, T.; Fujiwara, Y. Org. Lett. 1999,
1, 2097.
(18) Grimster, N. P.; Gauntlett, C.; Godfrey, C. R. A.; Gaunt, M.
J. Angew. Chem. Int. Ed. 2005, 44, 3125.
(19) Capito, E.; Brown, J. M.; Ricci, A. Chem. Commun. 2005,
1854.
1
(22) Compound 19: H NMR (400 MHz, CDCl3): d = 7.77 (1 H,
m, indole H-4 or H-7), 7.76 (1 H, m, indole H-4 or H-7), 7.53
(1 H, s, indole H-2), 7.21 (1 H, m, indole H-5 or H-6), 7.13
(1 H, m, indole H-5 or H-6), 4.95 (1 H, dd, J = 7.0, 6.0 Hz,
acetonide CH), 3.83 (1 H, dd, J = 8.9, 7.0 Hz, acetonide
CH2), 3.69 (1 H, dd, J = 8.9, 6.0 Hz, acetonide CH2), 1.87 (3
H, s, CH3), 1.79 (3 H, s, CH3), 1.45 (3 H, m, acetonide CH3),
1.37 (3 H, m, acetonide CH3). 13C NMR (100 MHz, CDCl3)
d = 135.7, 129.1, 123.6, 121.3, 120.5, 119.2, 113.6, 110.1,
109.0, 79.5, 65.4, 60.3, 26.0, 25.1, 24.8, 23.2. MS (ES):
m/z = 517 [M + H]+. Anal. Calcd for C32H40N2O4 (%): C,
74.39; H, 7.80; N, 5.42. Found: C, 74.37; H, 7.77; N, 5.39.
(23) Procedure for PdCl2-Catalyzed Oxidative Heck
Coupling (Table 2, Entry 5).
A mixture of indole 13 (78 mg, 0.30 mmol), PdCl2 (15 mg,
0.085 mmol), Cu(OAc)2 (0.168 g, 0.92 mmol) and ethyl
acrylate (0.10 mL, 0.92 mmol) in dry MeCN (3.5 mL) was
deoxygenated and heated at 40 °C in a capped Schlenk tube.
After 24 h, the reaction vessel was cooled to r.t. Brine was
added (25 mL) and the mixture was extracted with EtOAc
(3 × 25 mL). The combined organic phases were dried over
MgSO4, filtered and concentrated in vacuo. Purification by
flash chromatography (gradient elution with PE–Et2O
mixtures from 8:2 to 4:6) afforded product 12 (92 mg, 86%
yield).
Synlett 2006, No. 9, 1319–1322 © Thieme Stuttgart · New York