Total synthesis of a keramamide

Article abstract of DOI:10.1039/a901928f

The first total synthesis of a molecule possessing the structure proposed for keramamide J is described, providing data indicating that the structure of this natural product should be re-examined.

Full text of DOI:10.1039/a901928f

Total synthesis of a keramamide
Jennifer A. Sowinski and Peter L. Toogood*
Willard H. Dow Laboratory, Department of Chemistry, University of Michigan, Ann Arbor, MI 48109-1055, USA.
E-mail: peter.toogood@wl.com
Received (in Corvallis, OR, USA) 3rd March 1999, Accepted 15th April 1999
NHBoc
OMe
NHBoc
OMe
The first total synthesis of a molecule possessing the
structure proposed for keramamide J is described, providing
data indicating that the structure of this natural product
should be re-examined.
N
N
i–iv
v,vi
EtO2C
EtO2C
4
S
S
7
8
Scheme 2 Reagents and conditions: i, NaOH, MeOH; ii, BOP, Et3N,
HN(Me)OMe; iii, LiAlH4; iv, Ph3PNCHCO2Et (70% for 4 steps); v, NaOH,
MeOH; vi, l-Trp-OAllyl, Ph2POCl (68% for 2 steps).
Keramamide J (KJ; 1)1 is the simplest member of a class of
Theonella natural products that collectively exhibit cyto-
toxic,123 anti-fungal4 and anti-oxidant activity.5 To study the
biological activities of these molecules, we required a general
synthetic route that would be amenable to the synthesis of all
members of this series.4 Herein, we report our progress towards
this goal, including the first total synthesis of a keramamide
possessing structure 1.
i,ii
iii–vii
3
CbzHN
CbzHN
O
OH
10
9
Scheme 3 Reagents and conditions: i, NaBH4, CeCl3; ii, separate (76% for
2 steps); iii, SEMCl, 2,6-lutidine; iv, O3, DMS; v, NaClO2, isobutene; vi,
H2, Pd/C; vii, FmocOSu, aq. NaHCO3 (66% for 5 steps).
O
O
O
H
H
N
15
N
H
N
N
N
H
13
H
H
OH
O
O
O
HN
O
N
O
1
O
NH
prepared from l-isoleucine via displacement of the correspond-
ing Weinreb amide with vinyl magnesium bromide. Reduction
of this enone with NaBH4 under the Luche conditions9 provided
a 4 : 1 mixture of alcohols which could be separated by column
chromatography. Protection of the major product (10)10 as its
(trimethylsilyl)ethoxymethyl (SEM) acetal, followed by a two
step oxidation of the double bond, provided the carboxylic acid.
Hydrogenolytic cleavage of the Cbz group and subsequent
treatment with FmocOSu† gave compound 3. Deprotection of
fragment 4, followed by condensation with acid 3, gave the
corresponding tripeptide in 77% yield (Scheme 4). Subsequent
cleavage of the allyl ester and coupling to dipeptide 5 gave
linear precursor to the KJ macrocycle 11. Following deprotec-
tion of the C-terminus, and removal of the Fmoc group under
standard conditions, macrocyclization proceeded smoothly to
provide cyclic peptide 12 in 34–51% overall yields from
intermediate 11.
O
N
9
N
2
H
H
S
OMe
1
N
N
2
H
H
Our synthetic plan for constructing KJ is shown in Scheme
1.6 A convergent route was devised, employing four fragments
of comparable complexity (3–6). We decided to mask the
electrophilic keto-amide moiety7 as a protected a-hydroxy
carbonyl group until the conclusion of the synthesis to avoid
nucleophilic attack at this center. In addition, we elected to
attempt ring closure between the amine of fragment 3 and the C-
terminus of the alanine residue, which was anticipated to be less
prone to epimerization than the tryptophan residue and easier to
couple than the a-hydroxy acid. Late incorporation of side chain
fragment 6 was selected to increase convergency and minimize
potential side reactions during the critical macrocyclization
reaction.
To prepare the KJ side chain, (S)-glycidol was oxidized to
glycidic acid using RuCl3 and NaIO4,11 then coupled with l-
isoleucine methyl ester to give peptide 13 as a single
diastereomer (Scheme 5). Attack of the oxirane by azide ion at
Thiazole 78 was extended via reduction of the ester to an
aldehyde and Wittig olefination to introduce the E-double bond
(Scheme 2). Saponification of the ethyl ester, followed by
coupling to l-tryptophan allyl ester provided fragment 4. To
obtain the hexanoic acid fragment 3, enone 9 (Scheme 3) was
O
BocHN
OAllyl
NHFmoc
O
N
SEMO
H
O
i–iv
HN
O
4
O
HN
N
O
BocHN
H2N
OAllyl
N
H
N
H
Ar
O
O
5
S
OMe
11
H
FmocHN
SEMO
O
O
BocHN
N
Ar =
N
1
OH
H
O
H
N
N
O
O
H
H
v–vii
HN
O
O
SEMO
OH
OTES
O
NH
N
OAllyl
6
3
N
H
N
O
O
NHBoc
OMe
H
Ar
N
S
OMe
N
H
S
12
4
N
H
Scheme 4 Reagents and conditions: i, HCl, Et2O; ii, 3, DCC, HOBt,
Pri2EtN; iii, Pd(PPh3)4, dimedone; iv, 5, DCC, HOBt, Pri2EtN (56% for 4
steps); v, Pd(PPh3)4, dimedone; vi, Et2NH; vii, (PhO)2PON3, NaHCO3
(51% over 3 steps).
SEM = Me3SiCH2CH2OCH2
Scheme 1
Chem. Commun., 1999, 981–982
981

and 0.1 M NaOH (room temperature, overnight), followed by 6
M HCl (110 °C, 24 h), produced l-isoleucine containing less
than 10% d-allo-isoleucine by chiral HPLC analysis, support-
ing assignment of the l-configuration to carbon-13 in the
synthetic material.17 Upon standing in aqueous solution, the
synthetic material partially converted ( ~ 10%) to a new product
possessing 1H NMR resonances that match those observed for
KJ, consistent with the notion that these two molecules differ at
a single, epimerizable center.
O
O
i,ii
iii–vi
vii–ix
6
N3
N
CO2Bn
OH
N
CO2Me
H
O
O
H
OTES
13
14
Scheme 5 Reagents and conditions: i, RuCl3, NaIO4; ii, l-Ile-OMe, DCC,
HOBt (36% for 2 steps); iii, NaN3, MgSO4; iv, LiOH; v, CsCO3, BnBr; vi,
TESCl, imidazole (48% for 4 steps); vii, PPh3, H2O; viii, p-O2NC6H4-
OCHO; ix, H2, Pd/C (79% for 3 steps).
Based on the preceding analysis, we conclude that our
synthesis proceeded as intended to correctly provide a molecule
possessing structure 1. This work strongly indicates that the
structure of KJ should be revised. However, since no natural KJ
is presently available,15 structural re-assignment will require
either re-isolation or total synthesis of the correct structure.
This work was supported by NSF grant CHE-93221233, and
by the donors to the Petroleum Research Fund, administered by
the American Chemical Society.
the less substituted position, followed by transesterification and
protection of the hydroxy function produced azido peptide 14.
Staudinger reduction of the azide to a primary amine,12
followed by formylation with p-nitrophenyl formate and
hydrogenolytic cleavage of the ester, provided fragment 6.
Treatment of macrocycle 12 with HCl in Et2O–MeOH–
CHCl3 deprotected the amine and secondary alcohol, and
fragment 6 was attached using HATU† to produce alcohol 15
(Scheme 6).13 Oxidation of the alcohol was performed under
mild, non-acid conditions, using IBX† in DMSO.14 Finally,
removal of the TES group was achieved by stirring the peptide
over Amberlite IR-120 suspended in EtOAc. Purification of the
final product by silica gel chromatography followed by
Notes and references
† Abbreviations: HATU = O-(7-azabenzotriazol-1-yl)-1,1,3,3-tetramethyl-
uronium hexafluorophosphate, IBX = 1-hydroxy-1,3-dihydro-1,2-benzio-
doxol-3-one 1-oxide, BOP
= benzotriazol-1-yloxytris(dimethylamino)-
1
reversed-phase HPLC provided material that exhibits H, 13C,
phosphonium hexafluorophosphate, FmocOSu = N-(fluoren-9-ylmethox-
ycarbonyloxy)succinimide, HOBt = 1-hydroxybenzotriazole hydrate.
1H-1H COSY and inverse detected 1H–13C HMQC NMR
spectra consistent with the proposed structure 1, and a high
resolution mass spectrum corresponding to the expected
molecular formula.
1 J. Kobayashi, F. Itagaki, H. Shigemori, T. Takao and Y. Shimonishi,
Tetrahedron, 1995, 51, 2525.
2 F. Itagaki, H. Shigemori, M. Ishibashi, T. Nakamura, T. Sasaki, and J.
Kobayashi, J. Am. Chem. Soc., 1992, 57, 5540.
3 N. Fusetani, T. Sugawara, S. Matsunaga and H. Hirota, J. Am. Chem.
Soc. 1991, 113, 7811.
4 S. P. Gunasekara, S. A. Pomponi and P. J. McCarthy, J. Nat. Prod.,
1994, 57, 79.
5 J. Kobayashi, F. Itagaki, H. Shigemori, M. Ishibashi, K. Takahashi, M.
Ogura, S. Nagasawa, T. Nakamura, H. Hirota, T. Ohta and S. Nozoe,
J. Am. Chem. Soc., 1991, 113, 7812.
6 Preliminary work on this synthesis has been published: J. A. Sowinski
and P. L. Toogood, Tetrahedron Lett., 1995, 36, 67.
7 Examples of other ketoamide containing peptides see: N. Fusetani, S.
Matsunaga, H. Matsumoto and Y. Y. Takebayashi, J. Am. Chem. Soc.,
1990, 112, 7053; M. Hagihara and S. L. Schreiber, J. Am. Chem. Soc.,
1992, 114, 6570; S. Toda, C. Kotake, T. Tsuno, Y. Narita, T. Yamasaki
and M. Konishi, J. Antibiot., 1992, 45, 1580; T. Aoyagi, M. Nagai, K.
Ogawa, F. Kojima, M. Okada, T. Ikeda, M. Hamada and T. Takeuchi,
J. Antibiot., 1991, 44, 949; M. Nagai, K. Ogawa, Y. Muraoka, H.
Naganawa, T. Aoyagi and T. Takeuchi, J. Antibiot., 1991, 44, 956.
8 J. A. Sowinski and P. L. Toogood, J. Org. Chem., 1996, 61, 7671.
9 J.-L. Luche, J. Am. Chem. Soc., 1978, 100, 2226; J.-L. Luche, L.
Rodriguez-Hahn and P. J. Crabbé, J. Chem. Soc., Chem. Commun.,
1978, 601.
10 The major product from this reduction was identified through its
conversion to the corresponding oxazolidinone (NaH, DMF) and 1H
NMR spectral comparison with related literature compounds, See for
example, T. Ibuka, H. Habashita, A. Otaka, N. Fujii, Y. Oguchi, T.
Uyegara and Y. Yamamoto, J. Org. Chem., 1991, 56, 4370.
11 C. H. Behrens and K. B. Sharpless, J. Org. Chem., 1985, 50, 5696.
12 H. Staudinger and J. Meyer, Helv. Chim. Acta, 1919, 2, 635; N. Knouzi,
M. Voultier and R. Carrie, Bull. Soc. Chim. Fr., 1985, 815.
13 L. A. Carpino, J. Am. Chem. Soc., 1993, 115, 4397; L. A. Carpino and
A. El-Faham, J. Org. Chem., 1994, 59, 695; L. A. Carpino, A. El-
Faham, C. A. Minor and F. Albericio, J. Chem. Soc., Chem. Commun.,
1994, 201; L. A. Carpino, A. El-Faham and F. Albericio, Tetrahedron
Lett., 1994, 35, 2279; L. A. Carpino and A. El-Faham, J. Org. Chem.,
1995, 60, 3561.
O
O
O
H
H
N
N
H
N
N
N
H
H
H
O
O
O
OTES
HN
HO
N
i,ii
iii,iv
O
O
NH
12
1
N
H
Ar
S
OMe
15
Scheme 6 Reagents and conditions: i, HCl, MeOH; ii, 6, HATU,
2,4,6-collidine (48% for 2 steps); iii, IBX, DMSO; iv, Amberlite IR-120
(95% for 2 steps).
From the NMR data, it is apparent that our synthetic
keramamide is not the same compound reported by Kobayashi
and co-workers.1,15 Comparison of the NMR spectra leads us to
conclude that this synthetic keramamide and KJ are configura-
tional isomers. In support of the assignment of structure 1 to the
synthetic product, a very close correlation is observed between
the spectral data for this compound and the data published for
KF (2; Table 1), which differs only in the replacement of the
tryptophan in structure 1 by Z-didehydrotryptophan. In partic-
1
ular, the H and 13C chemical shifts at C-9 and C-13 for these
two compounds are in excellent agreement (1H Dd @ 0.06, 13
C
Dd @ 1.3 ppm). In contrast, the published data for KJ more
closely resemble the data for keramamide G (KG) which is
epimeric to KF at carbon-13.1,16 We note that the degradation
conditions used by Kobayashi to determine the absolute
configuration at carbon-13 in KJ have been found previously to
cause epimerization at this center in a closely related molecule
and possibly could have been misleading.3 Degradation of the
synthetic keramamide under milder conditions using 30% H2O2
Table 1 Selected 1H and 13C NMR resonances, and optical rotations
published for keramamides F, G, J and observed for compound 1
14 M. Frigerio, M. Santagostino, S. Sputore and G. Palmisano, J. Org.
Chem., 1995, 60, 7272.
15 Authentic samples of KJ and KF are not available. We thank Professor
Kobayashi for providing a copy of the 1H NMR spectrum for natural
KJ.
16 Chemical degradation of KG converts the homo-Ile fragment to d-
isoleucine indicating that KG possesses the (R) configuration at both C-
13 and C-15.
17 A peak corresponding to l-alanine was also detected in this analysis,
indicating that this residue did not epimerize during the cyclization
reaction.
dH
H-9
dC
[a]2D5
a
H-13
C-9
C-13
Keramamide G 4.81
5.49
5.49
5.19
5.25
53.7
53.7
51.4
51.7
56.7
56.1
61.0
59.7
+10.0
+8.4
210.0
Keramamide J
Compound 1
4.75
5.31
Keramamide F 5.33
225
a Given in units of 1021 degrees cm2 g21
.
Communication 9/01928F
982
Chem. Commun., 1999, 981–982