Rapid, high-yield, solid-phase synthesis of the antitumor antibiotic sansalvamide a using a side-chain-tethered phenylalanine building block

Article abstract of DOI:10.1021/ol0002830

(matrix presented) A 10-step solid-phase synthesis of the cytotoxic depsipeptide sansalvamide A (1) has been accomplished in an overall yield of 67% with >95% purity employing polymer-bound phenylalanine building block 2. Both the N- and C-termini of 2 are extended followed by on-resin head-to-tail macrocyclization of the linear peptide in a high yield. This should be a general stategy for the synthesis of diverse libraries of cyclic peptides and depsipeptides that contain exclusively phenylalanine and other hydrophobic side chains.

Full text of DOI:10.1021/ol0002830

ORGANIC
LETTERS
2000
Vol. 2, No. 23
3743-3746
Rapid, High-Yield, Solid-Phase Synthesis
of the Antitumor Antibiotic
Sansalvamide A Using a
Side-Chain-Tethered Phenylalanine
Building Block
Younghee Lee and Richard B. Silverman*
Department of Chemistry, Northwestern UniVersity, EVanston, Illinois 60208-3113
agman@chem.northwestern.edu
Received September 26, 2000
ABSTRACT
A 10-step solid-phase synthesis of the cytotoxic depsipeptide sansalvamide A (1) has been accomplished in an overall yield of 67% with >95%
purity employing polymer-bound phenylalanine building block 2. Both the N- and C-termini of 2 are extended followed by on-resin head-to-tail
macrocyclization of the linear peptide in a high yield. This should be a general stategy for the synthesis of diverse libraries of cyclic peptides
and depsipeptides that contain exclusively phenylalanine and other hydrophobic side chains.
Sansalvamide A (1) is a cyclic depsipeptide produced by a
marine fungus of the genus Fusarium found on Little San
Salvador Island, Bahamas.1 Structurally, sansalvamide A is
of the amino acids after acid hydrolysis. This highly
lipophilic natural product was found to have significant
cancer cell cytotoxicity with a mean IC50 value of 27.4 µg/
mL against the National Cancer Institute’s 60 cell-line panel.
Therefore, an efficient synthetic methodology to sansalva-
mide A, which would allow extensive diversity for biological
screening purposes, is highly desirable.
Many of the naturally occurring pseudopeptides, such as
Dolastin 10 and 15,2 and cyclic peptides3 are hydrophobic
in nature and pharmacologically active. Although the precise
correlation between biological activities and hydrophobicity
is unknown, at least one phenylalanine or modified phenyl-
(1) Belofsky, G. N.; Jensen, P. R.; Fenical, W. Tetrahedron Lett. 1999,
40, 2913-2916.
composed of four hydrophobic amino acids (Phe, 2 Leu, Val)
and one hydrophobic hydroxy acid ((S)-2-hydroxy-4-meth-
ylpentanoic acid; O-Leu), with five stereogenic centers all
having S-stereochemistry, as determined by an extensive
NMR analysis and chiral capillary GC of the ester derivatives
(2) (a) Poncet, J.; Hortala, L.; Busquet, M.; Gueritt-Voegelein, F.; Thoret,
S.; Pierre, A.; Atassi, G.; Jouin, P. Bioorg. Med. Chem. Lett. 1998, 8, 2855-
2858. (b) Akaji, K.; Hayashi, Y.; Kiso, Y.; Kuriyama, N. J. Org. Chem.
1999, 64, 405-411. (c) Pettit, G. R.; Kamano, Y.; Herald, C. L.; Fujii, Y.;
Kizu, H.; Boyd, M. R.; Boettner, F. E.; Doubek, D. L.; Schmidt, J. M.;
Chapuis, J. C. Tetrahedron 1993, 49, 9150-9170.
10.1021/ol0002830 CCC: $19.00 © 2000 American Chemical Society
Published on Web 10/21/2000

Scheme 1. Synthesis of Phenylalanine Building Block 2a
a Reagents and conditions: (a) n-BuLi, THF, -78 °C, then allylchlorodimethylsilane, 82%; (b) t-BuLi, THF, -78 °C, then DMF, 82%;
(c) methyl 2-acetamido-2-(dimethoxyphosphinyl)acetate, tetramethylguanidine, THF, -78 °C to room temperature, 86%; (d) [((S,S)-Et-
DuPHOS)-Rh]+ (1 mol %), H2 (1 atm), CH2Cl2, 23 h, 100%; (e) (Boc)2O, DMAP (cat.), THF, reflux, 1 h; (f) hydrazine, MeOH, 4 h, 93%
for two steps; (g) 9-BBN, THF, room temperature, 5 h, then 6, DMF, Pd(PPh3)4, 2 N Na2CO3, 75 °C, 3 days; (h) Br2 in CH2Cl2, 20 min;
(i) 2% thioanisole and 50% TFA in CH2Cl2, 15 min; (j) (R)-MTPACl (3 equiv), DIPEA (4 equiv), CH2Cl2, 4 h.
alanine residue is commonly found in these compounds.
Given the role of phenylalanine as an important pharmaco-
phore in many biologically active agents, solid-phase pro-
tocols to phenylalanine-containing compounds are highly
desirable.
yield synthesis of libraries of cyclic depsipeptides and cyclic
peptides containing hydrophobic side chains exclusively.
Our original synthetic method for the polymer-bound
phenylalanine building block 2 involves a C-C bond-
forming reaction of lithiated Schollkopf’s bislactim ether with
4-allyldimethylsilylbenzyl bromide as a key step which
results in the formation of two diastereomers in a ratio of
94 to 6.5 The tedious column chromatography procedure
required for the separation of the major isomer led us to
investigate an alternative synthetic route for the silylated
phenylalanine building block 2 (Scheme 1).
Lithiation of 1,4-dibromobenzene with n-butyllithium
followed by treatment with allylchlorodimethylsilane gave
1-allyldimethylsilyl-4-bromobenzene; further lithiation with
tert-butyllithium and reaction with DMF produced 4-allyl-
dimethylsilylbenzaldehyde (3). Tetramethylguanidine-medi-
ated condensation of 3 with Horner-Emmons reagent
afforded a (Z)-enamide ester (4) exclusively.7 Subsequent
Recently, we reported arylsilane-based “traceless linker”
strategies4 to attach the aromatic side chain of phenylalanine5
and â-homophenylalanine6 to a polystyrene resin, based on
the earlier work of Plunkett and Ellman.4a Our work
accommodates the elaboration of solid-phase bound aromatic
amino acid residues in both the N- and C-terminal directions,
which serves as an efficient synthetic tool for constructing
libraries of peptides and peptidomimetics containing exclu-
sively phenylalanine and other hydrophobic side chains. The
synthesis of cyclic peptides and depsipeptides is generally
hampered by low yields in the cyclization step, which
requires high dilution conditions and is accompanied by
dimerization and oligomerization side reactions; in this case,
not only are yields low but purification becomes tedious.
Here we report an improved synthetic procedure for the
polymer-bound phenylalanine building block 2 and its use
(3) (a) Hawkins, G. J.; Lavin, M. F.; Marshall, K. A.; van den Brenk,
A. L.; Watters, D. J. J. Med. Chem. 1990, 33, 1634-1638. (b) Degnan, B.
M.; Hawkins, C. J.; Lavin, M. F.; McCaffrey, E. J.; Parry, D. L.; van den
Brenk, A. L.; Watters, D. J. J. Med. Chem. 1989, 32, 1349-1354. (c) Norley,
M. C.; Pattenden, G. Tetrahedron Lett. 1998, 39, 3087-3090. (d) Freeman,
D. J.; Pattenden, G. Tetrahedron Lett. 1998, 39, 3251-3254. (e) McKeever,
B.; Pattenden, G. Tetrahedron Lett. 1999, 40, 9317.
(4) (a) Plunkett, M. J.; Ellman, J. A. J. Org. Chem. 1995, 60, 6006-
6007. (b) Plunkett, M. J.; Ellman, J. A. J. Org. Chem. 1997, 62, 2885-
2893. (c) Chenera, B.; Finkelstein, J. A.; Veber, D. F. J. Am. Chem. Soc.
1995, 117, 11999-12000. (d) Boehm, T. L.; Showalter, H. D. H. J. Org.
Chem. 1996, 61, 6498-6499. (e) Woolard, F. X.; Paetsch, J.; Ellman, J. A.
J. Org. Chem. 1997, 62, 6102-6103. (f) Hu, Y.; Porco, J. A. Jr.; Labadie,
J. W.; Gooding, O. W.; Trost, B. M. J. Org. Chem. 1998, 63, 4518-4521.
(g) Newlander, K. A.; Chenera, B.; Veber, D. F.; Yim, N. C. F.; Moore,
M. L. J. Org. Chem. 1997, 62, 6726-6732. (h) Han, Y.; Walker, S. D.;
Young, R. N. Tetrahedron Lett. 1996, 37, 2703-2706.
in the first synthesis of the hydrophobic cyclic depsipeptide
antitumor antibiotic sansalvamide A. The methodology
reported here should prove to be general for the rapid, high-
(5) Lee, Y.; Silverman, R. B. J. Am. Chem. Soc. 1999, 121, 8407-8408.
(6) Lee, Y.; Silverman, R. B. Org. Lett. 2000, 2, 303-306.
3744
Org. Lett., Vol. 2, No. 23, 2000

Scheme 2. Synthesis of Sansalvamide Aa
a Reagents and conditions: (a) LiOH (5 equiv), THF/H2O (7:1), room temperature, 16 h; (b) HBTU (4 equiv), DIPEA (4 equiv), 9 (4
equiv), NMP, 16 h; (c) 2% thioanisole and 50% TFA in CH2Cl2, 15 min; (d) Fmoc-Leu-OH (4 equiv), HBTU (4 equiv), DIPEA (4 equiv),
NMP, 6 h; (e) 20% piperidine in DMF, 40 min; (f) Fmoc-Val-OH (4 equiv), HBTU (4 equiv), DIPEA (4 equiv), NMP, 6 h; (g) CHCl3/
AcOH/NMM (37:2:1), Pd(PPh3)4 (4 equiv), 3 h; (h) HBTU (4 equiv), DIPEA (4 equiv), NMP, 16 h; (i) 2% thioanisole and 50% TFA in
CH2Cl2, 36 h; (j) allyl bromide, K2CO3, acetone, 48 h, 95%; (k) Boc-Leu-OH, DMAP (cat), DCC, CH2Cl2, 48 h, 63%; (l) 50% TFA in
CH2Cl2, 5 min.
asymmetric catalytic hydrogenation of 4 with the com-
mercially available [((S,S)-Et-DuPHOS)-Rh]+ catalyst system
in CH2Cl2 under H2 (1 atm) created the desired chiral
R-carbon in quantitative yield without affecting the terminal
olefin of the allyl group.8 Interchange of the N-acetyl group
for N-Boc was carried out in one pot by treatment of the
chiral intermediate with Boc2O in the presence of a catalytic
amount of DMAP in refluxing THF followed by hydrolysis
of the resulting mixed imide with excess hydrazine in MeOH
for 4 h. This provided 5 in 93% for two steps.9 Hydroboration
of the terminal olefin of 5, followed by Suzuki coupling of
the generated borane complex with 3-iodobenzamidometh-
ylpolystyrene (6),10 resulted in the production of resin-bound
phenylalanine building block 2. The loading level (0.12
mmol/g) of 2 was determined by the mass balance of 7,
which was obtained by cleaving 2 from the resin with
bromine in CH2Cl2.11 Treatment of 2 with 50% TFA in CH2-
Cl2 for 15 min, followed by reaction with (R)-(-)-R-
methoxy-R-(trifluoromethyl)phenylacetyl chloride (MTPACl)
in the presence of base and subsequent cleavage of the
product with bromine in CH2Cl2, produced Mosher amide
8.12 Analysis of both 1H and 19F NMR spectra of 8 indicated
high enantiopurity (>98% ee) of polymer-bound building
block 2.
As illustrated in Scheme 2, solid-phase synthesis of
sansalvamide A was initiated by deprotection of the ester
group of 2, followed by reaction of the resulting acid with
O-allyl Leu-O-Leu (9) under standard amide coupling
conditions (O-benzotriazol-1-yl-N,N,N′,N′-tetramethyluro-
nium hexafluorophosphate (HBTU), diisopropylethylamine
(DIPEA), and N-methylpyrrolidinone (NMP)), to give 10.
The depsipeptide ester 9 required for the synthesis of 10 was
prepared in three steps: (1) allylation of hydroxy acid 12
with allyl bromide and K2CO3 in acetone; (2) condensation
of 13 with Boc-Leu-OH in the presence of catalytic (di-
methylamino)pyridine (DMAP) and dicyclohexylcarbodii-
mide (DCC) in CH2Cl2 to give 14; (3) cleavage of the N-Boc
group with TFA. After deprotection of the N-Boc protecting
group and extension of the N-terminus using Fmoc strategy
in a three-step sequence (coupling to Fmoc-Leu-OH/depro-
tection of Fmoc/coupling to Fmoc-Val-OH), protected linear
intermediate 11 was produced. The allyl ester was removed
with Pd(0) in CHCl3/AcOH/N-methylmorpholine (NMM)
(37:2:1),13 which was followed by cleavage of the Fmoc
(7) (a) Schmidt, U.; Lieberknecht, A.; Wild, J. Synthesis 1984, 53-60.
(b) Schmidt, U.; Griesser, H.; Leitenberger, V.; Lieberknecht, A.; Mangold,
R.; Meyer, R.; Riedl, B. Synthesis 1992, 487-490.
(8) (a) Burk, M. J.; Feaster, J. E.; Nugent, W. A.; Harlow, R. L. J. Am.
Chem. Soc. 1993, 115, 10125-10138. (b) Burk, M. J.; Allen, J. G.; Kiesman,
W. F. J. Am. Chem. Soc. 1998, 120, 657-663.
(9) Burk, M. J.; Allen, J. G. J. Org. Chem. 1997, 62, 7054-7057.
(10) Prepared by amide coupling of 3-iodobenzoic acid with aminom-
ethylated polystyrene (1.02 mmol/g, 100-200 mesh, purchased from
Novabiochem Corp.) employing diisopropylcarbodiimide and HOBT in
DMF for 2 days at room temperature.
(11) Chan, T. H.; Fleming, I. Synthesis 1979, 761-786.
(12) Dale, J. A.; Mosher, H. S. J. Am. Chem. Soc. 1973, 95, 512-519.
Org. Lett., Vol. 2, No. 23, 2000
3745

Scheme 3. Alternative Synthesis of Intermediate 11 in the Synthesis of Sansalvamide Aa
a Reagents and conditions: (a) LiOH (5 equiv), THF/H2O (7:1), room temperature, 16 h; (b) O-allylLeu (4 equiv), HBTU (4 equiv),
DIPEA (4 equiv), NMP, 16 h; (c) 2% thioanisole and 50% TFA in CH2Cl2, 15 min; (d) Fmoc-Leu-OH (4 equiv), HBTU (4 equiv), DIPEA
(4 equiv), NMP, 6 h; (e) 20% piperidine in DMF, 40 min; (f) Fmoc-Val-OH (4 equiv), HBTU (4 equiv), DIPEA (4 equiv), NMP, 6 h; (g)
CHCl3/AcOH/NMM (37:2:1), Pd(PPh3)4 (4 equiv), 3 h; (h) 13, MSNT, NMI, CH2Cl2, 16 h.
group. Cyclization of the depsipeptide on the polymer support
was readily achieved with HBTU and DIPEA in NMP for
16 h. One important advantage of using a solid-phase strategy
is that cyclization of the linear (depsi)peptide can occur
without the need for dilute conditions and without linear
polymer byproduct formation. Finally, cleavage of the
cyclized product from the support was performed with 50%
TFA/CH2Cl2 for 36 h to release sansalvamide A, identical
by NMR spectral comparison to natural sansalvamide A,14
in a 67% yield based on the initial loading level of 2. The
lalanine building block 2 in an overall yield of 67% with
>95% purity. Although side-chain tethering of an amino acid
has been previously utilized for solid-phase cyclic peptide
synthesis, this method has been limited to amino acids having
a polar (heteroatomic) side chain which is attached to the
polymer support.13,16 The advantages of the solid-phase
approach, therefore, have been lost for those cyclic (depsi)-
peptides that contain only hydrophobic side chains. The
polymer-bound phenylalanine building block developed in
this study, however, can provide rapid access to cyclic
peptides and depsipeptides that consist exclusively of
hydrophobic side chains. An added advantage of the silicon-
based traceless linker strategy4a is that the arylsilyl linkage
used is compatible with a variety of reaction conditions, as
described in the present synthesis, and can be cleaved under
several different conditions, which is desirable for linkers
used in the synthesis of complex molecules. This methodol-
ogy should prove to be a general rapid and high-yield
approach to the synthesis of libraries of cyclic peptides and
depsipeptides containing exclusively phenylalanine and other
hydrophobic amino acids.
1
high purity (>95%) of the crude product determined by H
NMR spectral analysis indicated that the solid-phase meth-
odology proceeded efficiently.
Except for the solution synthesis of 9, the synthesis of
sansalvamide A shown in Scheme 2 is resin bound. To devise
a more efficient and general methodology which can be
utilized for automated solid-phase synthesis of sansalvamide
A analogues, on-resin ester bond formation was carried out
(Scheme 3). The resin-bound phenylalanine building block
2 was elaborated in both the C- and N-directions to give
resin-bound diprotected tetrapeptide 16. Deprotection of the
allyl group in 16 followed by coupling of the resulting
carboxylic acid with 13 in the presence of 1-(mesitylene-2-
sulfonyl)-3-nitro-1H-1,2,4-triazole (MSNT) and N-meth-
ylimidazole (NMI) in CH2Cl215 afforded 11, which again led
to the synthesis of sansalvamide A in comparable yield and
purity as described in Scheme 2.
Supporting Information Available: Complete experi-
mental details and product characterization. This material is
available free of charge via the Internet at http://pubs.acs.org.
OL0002830
In conclusion, a 10-step total synthesis of the antitumor
antibiotic sansalvamide A was carried out from the pheny-
(16) For example see (a) Sabatino, G.; Chelli, M.; Mazzucco, S.;
Ginanneschi, M.; Papini, A. M. Tetrahedron Lett. 1999, 40, 809-812. (b)
Rovero, P.; Quartara, L.; Fabbri, G. Tetrahedron Lett. 1991, 32, 2639-
2642. (c) Kates, S. A.; Sole, N. A.; Johnson, C. R.; Hudson, D.; Barany,
G.; Albericio, F. Tetrahedron Lett. 1993, 34, 1549-1552. (d) Sugiura, T.;
Hamada, Y.; Shioiri, T. Tetrahedron Lett. 1987, 28, 2251-2254. (e) Zhong,
H. M.; Greco, M. N.; Maryanoff, B. E. J. Org. Chem. 1997, 62, 9326-
9330. (f) Cabrele, C.; Langer, M.; Beck-Sickinger, A. G. J. Org. Chem.
1997, 64, 44353-4361.
(13) Trzeciak, A.; Bannwarth, W. Tetrahedron Lett. 1992, 33, 4557-
4560.
(14) A sample of sansalvamide A was kindly donated by Professor
William Fenical of Scripps Institution of Oceanography (University of
California, San Diego).
(15) Nielsen, J. Tetrahedron Lett. 1996, 37, 8439-8442.
3746
Org. Lett., Vol. 2, No. 23, 2000