(+)-Trienomycins A, B, C, and F and (+)-mycotrienins I and II: Relative and absolute stereochemistry

Article abstract of DOI:10.1021/ja961400i

The complete relative and absolute stereochemistries have been elucidated for the ansamycin antibiotics (+)-trienomycins A, B, and C and their potent antifungal congeners, the (+)-mycotrienins I and II. A new species, (+)-trienomycin F, has also been isol

Full text of DOI:10.1021/ja961400i

8308
J. Am. Chem. Soc. 1996, 118, 8308-8315
(+)-Trienomycins A, B, C, and F and (+)-Mycotrienins I and
II: Relative and Absolute Stereochemistry
Amos B. Smith, III,*,† John L. Wood,† Weichyun Wong,† Alexandra E. Gould,†
Carmelo J. Rizzo,† Joseph Barbosa,† Kanki Komiyama,‡ and Satoshi Oh mura*,‡
Contribution from the Department of Chemistry, Monell Chemical Senses Center,
and Laboratory for Research on the Structure of Matter, UniVersity of PennsylVania,
Philadelphia, PennsylVania 19104, and The Kitasato Institute and School of Pharmaceutical
Sciences, Kitasato UniVersity, Minato-ku, Tokyo 108, Japan
ReceiVed April 29, 1996X
Abstract: The complete relative and absolute stereochemistries have been elucidated for the ansamycin antibiotics
(+)-trienomycins A, B, and C and their potent antifungal congeners, the (+)-mycotrienins I and II. A new species,
(+)-trienomycin F, has also been isolated and characterized. In addition, an end-game synthetic strategy for the
trienomycins and mycotrienins has been developed.
In 1985 Umezawa and co-workers at the Kitasato Institute
(Tokyo) disclosed the isolation and planar structures of five new
ansamycin antibiotics produced by Streptomyces sp. No. 83-
16.1 These compounds, designated the (+)-trienomycins A-E
(1-5), exhibited strong in vitro cytotoxicity against HeLa S3
cells;2 (+)-trienomycin A (1), the most potent congener, was
also active against the L-5178Y murine leukemia cell line as
well as human PLC hepatoma cells (IC50 0.01 µg/mL).3
(+)-Mycotrienins I and II (6 and 7) and (+)-mycotrienols I
and II (8 and 9) had previously been obtained from the
fermentation broth of Streptomyces rishiriensis T-23,3,4 ac-
companied by (+)-trienomycin A as a minor component. In
contrast with the trienomycins, the mycotrienins displayed potent
antifungal activity. The ansatrienins A and B and three minor
components, (+)-ansatrienins A2, A3 and A4 (10-12), were
independently isolated from the culture broth of Streptomyces
collinus.5 Subsequent studies established that ansatrienins A
and B were identical to mycotrienins I and II.6
The N-acylated alanine side chains of the (+)-trienomycins
and (+)-mycotrienins were identified via degradation (vide
1
infra). Extensive H and 13C NMR analyses, augmented by
† University of Pennsylvania.
‡ Kitasato University.
X Abstract published in AdVance ACS Abstracts, August 1, 1996.
(1) (a) Funayama, S.; Okada, K.; Komiyama, K.; Umezawa, I. J. Antibiot.
1985, 38, 1107. (b) Funayama, S.; Okada, K.; Iwasaki, K.; Komiyama, K.;
Umezawa, I. J. Antibiot. 1985, 38, 1677. (c) Nomoto, H.; Katsumata, S.;
Takahashi, K.; Funayama, S.; Komiyama, K.; Umezawa, I.; Ohmura, S. J.
Antibiot. 1989, 42, 479.
(2) (a) Umezawa, I.; Funayama, S.; Okada, K.; Iwasaki, K.; Satoh, J.;
Masuda, K.; Komiyama, K. J. Antibiot. 1985, 38, 699. (b) Funayama, S.;
Anraku, Y.; Mita, A.; Yang, Z.-B.; Shibata, K.; Komiyama, K.; Umezawa,
I.; Ohmura, S. J. Antibiot. 1988, 41, 1223.
(3) Hiramoto, S.; Sugita, M.; Andoj, C.; Sasaki, T.; Furihata, K.; Seto,
H.; Ohtake, N. J. Antibiot. 1985, 38, 1103.
(4) (a) Sugita, M.; Natori, Y.; Sasaki, T.; Furihata, K.; Shimazu, A.; Seto,
H.; Ohtake, N. J. Antibiot. 1982, 35, 1460. (b) Sugita, M.; Sasaki, T.; Furihata,
K.; Seto, H.; Ohtake, N. J. Antibiot. 1982, 35, 1467. (c) Sugita, M.; Natori,
Y.; Sueda, N.; Furihata, K.; Seto, H.; Ohtake, N. J. Antibiot. 1982, 35, 1474.
(d) Sugita, M.; Furihata, K.; Seto, H.; Ohtake, N.; Sasaki, T. Agric. Biol.
Chem. 1982, 46, 1111.
(5) (a) Damberg, M.; Russ, P.; Zeeck, A. Tetrahedron Lett. 1982, 23,
59. (b) Lazar, G.; Za¨hner, H.; Damberg, M.; Zeeck, A. J. Antibiot. 1983,
36, 187.
(6) (a) Wu, T. S.; Duncan, J.; Tsao, S. W.; Chang, C. J.; Keller, P. J.;
Floss, H. G. J. Nat. Prod. 1987, 50, 108. (b) Casati, R.; Beale, J. M.; Floss,
H. G. J. Am. Chem. Soc. 1987, 109, 8102, and references cited therein.
chemical interconversions, revealed the planar structures of the
macrolactam rings, which differ only in the oxidation state of
the aryl moiety.1,3,4 However, neither these initial efforts nor
S0002-7863(96)01400-X CCC: $12.00 © 1996 American Chemical Society

Stereochemistry of Trienomycins and Mycotrienins
J. Am. Chem. Soc., Vol. 118, No. 35, 1996 8309
extensive biosynthetic investigations6 have addressed the stereo-
chemistry of the macrocyclic skeleton.
Scheme 2
As a prelude to total synthesis,7 we have elucidated the
complete relative and absolute configurations for (+)-trieno-
mycins A-C (1-3)8 and (+)-mycotrienins I and II (6 and 7).9
These studies also led to the discovery and characterization of
a new congener, (+)-trienomycin F (13),10 as well as the
development of an end-game synthetic strategy for elaboration
of the trienomycins from a common advanced intermediate, (+)-
trienomycinol (14).10 Herein we provide a full account of this
venture.
A Degradation Strategy for (+)-Trienomycin A. With
ample quantities of the trienomycins available, we envisioned
the development of a degradation scheme beginning with
trienomycinol (14) and focusing on the three contiguous
stereogenic centers at C(11-13) and the isolated methoxy center
at C(3). Side-chain cleavages of individual trienomycins and
mycotrienins have been described previously (Scheme 1).
Umezawa isolated D-alanine upon acidic hydrolysis of (+)-
trienomycin A (1) (6 N HCl, 120 °C, 16 h).1b Earlier, Seto
liberated an intact N-acyl alanine moiety as its methyl ester via
alkaline methanolysis (NaHCO3, MeOH) of (+)-mycotrienin
II.4b Comparison of the latter with an authentic sample prepared
from L-alanine and cyclohexanecarbonyl chloride led to unam-
biguous assignment of the D configuration for the alanine
subunit.
In a model study, we explored the viability of our proposed
sequence for isolation of γ-lactone 18. Both enantiomers of
lactone 1811 were prepared via known methods from (+)- and
Scheme 1
(-)-malic acid.12 The model aromatic amide 19 was obtained
by coupling of lactone 18 with the Weinreb aluminum amide
derived from m-anisidine (Scheme 3).13 As expected, either
ozonolysis of 19 followed by reduction and acidification or
simple acidification furnished the desired lactone 18.
Scheme 3
Our point of departure (Scheme 2) was the known deacylation
of (+)-trienomycin A (1) to (+)-trienomycinol (14) with
LiAlH4,2b followed by acetonide formation to give (+)-15.
Initially we envisioned that a reductive ozonolysis sequence
would then provide the C(11-13) keto aldehyde 16 as well as
a mixture of products containing the C(3) stereocenter (i.e., 17).
Reduction of the aldehyde moiety in 17 and acid-promoted
cyclization were then expected to provide γ-lactone 18.
Turning next to the degradation of 1, ozonolysis14 of the
acetonide of trienomycinol (+)-15 and reductive workup with
(12) Preparation of hydroxy γ-lactone: (a) Saito, S.; Hasegawa, T.; Inaba,
M.; Nishida, R.; Fujii, T.; Nomizu, S.; Moriwake, T. Chem. Lett. 1984,
1389. (b) Tanaka, A.; Yamashita, K. Synthesis 1987, 570. Conversion to
O-methyl derivative: (c) Lardon, A.; Reichstein, T. HelV. Chim. Acta 1949,
32, 2003.
(13) (a) Levin, J. I.; Turos, E.; Weinreb, S. M. Synth. Commun. 1982,
12, 989. (b) Basha, A.; Lipton, M. F.; Weinreb, S. M. Tetrahedron Lett.
1977, 4171. (c) Lipton, M. F.; Basha, A.; Weinreb, S. M. Org. Synth. 1979,
59, 49.
(7) Smith, A. B., III; Barbosa, J.; Wong, W.; Wood, J. L. J. Am. Chem.
Soc. 1995, 117, 10777.
(8) Smith, A. B., III; Wood, J. L.; Wong, W.; Gould, A. E.; Rizzo, C.
J.; Funayama, S.; Ohmura, S. J. Am. Chem. Soc. 1990, 112, 7425.
(9) Smith, A. B., III; Wood, J. L.; Ohmura, S. Tetrahedron Lett. 1991,
32, 841.
(10) Smith, A. B., III; Wood, J. L.; Gould, A. E.; Ohmura, S.; Komiyama,
K. Tetrahedron Lett. 1991, 32, 1627.
(11) Doyle, M. P.; Van Oeveren, A.; Westrum, L. J.; Protopopova, M.
N.; Clayton, T. W., Jr. J. Am. Chem. Soc. 1991, 113, 8982.
(14) For a review of ozonolysis, see: Bailey, P. S. Ozonation in Organic
Chemistry; Academic Press: New York, 1978; Vol. 1, Chapter 4, p 15.

8310 J. Am. Chem. Soc., Vol. 118, No. 35, 1996
Smith et al.
Table 2. Calculated Energies and Coupling Constants for the
Me2S afforded keto aldehyde (+)-16 [[R]25 +45° (c 0.53,
D
Favored Monte Carlo Conformations of 16a-d
CHCl3)] in 35% yield along with a mixture of more polar
materials (Scheme 4). Reduction of the mixture with excess
NaBH4 followed by exposure to strong acid provided only a
trace (<1 mg) of γ-lactone 18. Moreover, the optical rotation
was only +0.01°, providing at best a tentative assignment of
the R configuration at C(3).
Scheme 4
MM2 energy
(kcal/mol)
ring
J11,12 (Hz) J12,13 (Hz) conformation
isomer
16a
-7.96
-7.89
-7.82
-7.61
-7.05
-9.95
-9.70
-9.13
-9.04
-11.05
-11.14
-10.98
-8.43
-8.29
-7.89
-7.78
-7.49
10.7
10.7
10.7
10.7
8.7
2.7
2.7
2.7
2.7
10.6
10.6
10.7
2.6
5.3
5.3
5.3
5.3
5.3
2.1
2.8
2.8
2.7
11.0
11.1
11.1
0.6
0.6
0.6
9.5
9.5
chair
chair
chair
chair
twist boat
chair
chair
chair
chair
chair
chair
chair
chair
chair
chair
twist boat
twist boat
16b
16c
16d
Relative and Absolute Stereochemistry of Fragment 16:
Molecular Modeling and Asymmetric Synthesis. Given a
viable degradation route to keto aldehyde (+)-16 we focused
on determination of the relative and absolute configurations of
C(11), C(12), and C(13). We first analyzed the relative
stereochemistry via 1H NMR, anticipating that the initial
assignments would ultimately be secured by asymmetric syn-
thesis. Importantly, the C(11,12) and C(12,13) coupling
constants for both (+)-15 and (+)-16 were quite similar to those
reported for the pentamethyl-1,3-dioxane 2015 (Table 1).
2.7
2.6
5.3
5.4
exptl: 15,16
8.5, 7.7
5.9, 5.6
constants (Table 2) for all conformations within 1.0 kcal/mol
of the global minima. By far the best correlation with the
experimental couplings was obtained for a twist-boat conformer
of the C(11,12)-anti, C(12,13)-syn isomer 16a [JC(11,12) ) 8.7
Hz, JC(12,13) ) 5.3 Hz], providing further support for our tentative
assignment.
Table 1. Observed J11,12 and J12,13 Values (Trienomycin
Numbering) for (+)-15, (+)-16, and 20
Reasonably confident of the relative stereochemistry, we
embarked on an asymmetric synthesis of 16a, the C(11,12)-
anti, C(12,13)-syn diastereomers (Scheme 5).18 DIBAL reduc-
tion of the known butenolide (+)-2119 afforded a 1:1 mixture
of epimeric lactols 22 (99% yield). Wittig methylenation (n-
BuLi, Ph3PCH3Br, THF, 0 °C; 83%), protection of the resultant
diol (+)-23 as the acetonide (98%), and desilylation (TBAF,
THF; quantitative) provided alcohol (-)-25. Treatment of (-)-
25 with excess m-CPBA and NaHCO3 (CH2Cl2 at reflux) then
furnished epoxide 26 as a 3:1 diastereomeric mixture in 81%
yield. To facilitate characterization of more advanced inter-
mediates, the isomers were separated chromatographically and
the major epimer [(-)-26] was subjected to Swern oxidation,
affording aldehyde (-)-27 in 97% yield. Standard Wittig
olefination (n-BuLi, Ph3PCH3Br, THF) proved unsatisfactory,
but a salt-free protocol (KH, Ph3PCH3Br)20 delivered the
requisite olefin (-)-28 in 70% yield. Hydroboration and
oxidative workup afforded alcohol (-)-29 (48%). Finally,
LAH-mediated epoxide opening followed by Swern oxidation
of the resultant alcohol gave keto aldehyde (-)-16a [[R]D25
-44.6° (c 0.53, CHCl3)] in 35% yield for the two steps. This
compound
J11,12 (Hz)
J12,13 (Hz)
(+)-15
(+)-16
20
8.5
7.7
7.9
5.9
5.6
5.3
The relative stereochemistry was further evaluated by subject-
ing the four diastereomers of keto aldehyde (+)-16 (a-d, Table
2) to a Monte Carlo conformational search.16 MacroModel17
was then employed to predict C(11,12) and C(12,13) coupling
(15) Pihlaja, K.; Kellie, G. M.; Riddell, F. G. J. Chem. Soc., Perkin Trans.
2 1972, 252.
(16) (a) MacroModel V2.5: Still, W. C.; Mohamadi, F.; Richards, N.
G. J.; Guida, W. C.; Lipton, M.; Liskamp, R.; Chang, G.; Hendrickson, T.;
DeGunst, F.; Hasel, W. Department of Chemistry, Columbia University,
New York, NY 10027. (b) Chang, G.; Guida, W. C.; Still, W. C. J. Am.
Chem. Soc. 1989, 111, 4379.
(18) Wong, W. Ph.D. Thesis, University of Pennsylvania, Philadelphia,
PA, 1993.
(19) (a) Hanessian, S.; Murray, P. J. Tetrahedron 1987, 43, 5055. (b)
Hanessian, S.; Murray, P. J. J. Org. Chem. 1987, 52, 1170.
(20) Trost, B. M.; Latimer, L. H. J. Org. Chem. 1978, 43, 1031.
(17) Coupling constants were determined in the NMR submode of
MacroModel.

Stereochemistry of Trienomycins and Mycotrienins
J. Am. Chem. Soc., Vol. 118, No. 35, 1996 8311
Scheme 5
Scheme 6
Following derivatization of 30 as the bis(Mosher ester) (+)-
32, careful comparison (i.e., GC/MS, HRMS, 1H and 13C INEPT
NMR) with authentic samples of the diastereomers (+)-32 and
(+)-33 unambiguously verified the R configuration at C(3)
(Scheme 7), in accord with our tentative earlier assignment. The
relative and absolute stereochemistry of the (+)-trienomycin A
skeleton thus proved to be 3R, 11S, 12R, 13R,23 completely
defining the structure of the natural product in conjunction with
the side-chain degradation studies of Umezawa.
Scheme 7
material proved identical to 16 obtained via degradation of (+)-1
in all respects except for the sign of optical rotation. Thus, the
relative stereochemistry was confirmed and the absolute stereo-
chemistry established as 11S, 12S, 13R. As described in the
accompanying paper,21 we have also prepared the enantiomer
(+)-16 as a key intermediate in a total synthesis of (+)-
trienomycins A and F.
An Effective Tactic for Determination of the C(3) Absolute
Configuration. Although our initial degradation scheme suc-
ceeded admirably in elucidation of the C(11-13) configurations,
the stereochemistry at C(3), the final stereocenter within the
trienomycin A macrocyclic skeleton, remained uncertain. We
envisioned that 2-methoxy-1,4-butanediol (30), an alternative
degradation target, could be generated via Fukuyama reduction
of the BOC-protected secondary amide 31, followed by ozon-
olysis and reduction of the derived hydroxy aldehyde. To this
end, (+)-1 was converted to the tris-BOC derivative (+)-31
(Scheme 6).22 Direct reduction of the amide with DIBAL or
LAH afforded products devoid of the C(3) methoxy group,
which presumably underwent â-elimination. However, treat-
ment of (+)-31 with ozone prior to LAH reduction provided
the desired diol 30 in 70% yield.
Relative and Absolute Configurations of (+)-Trienomycins
B and C. We next sought to elucidate the stereochemistries of
(+)-trienomycins B (2) and C (3) via chemical correlation. All
three congeners were believed to embody a common macro-
lactam, but in contrast with (+)-trienomycin A [(+)-1], the
absolute configurations of the alanine side chains in (+)-2 and
(+)-3 were not assigned in the initial structure work.1b
(21) See following article in this issue: Smith, A. B., III; Barbosa, J.;
Wong, W.; Wood, J. L. J. Am. Chem. Soc. 1996, 118, 8316.
(22) Fukuyama, T.; Nunes, J. J. J. Am. Chem. Soc. 1988, 110, 5196.
Also see: Flynn, D. L.; Zelle, R. E.; Grieco, P. A. J. Org. Chem. 1983, 48,
2424.
(23) Following the CIP sequence rules, the analogous configuration of
(+)-16 is designated 11S, 12S, 13R.

8312 J. Am. Chem. Soc., Vol. 118, No. 35, 1996
Smith et al.
Reverse-phase HPLC afforded pure samples of (+)-2 and
(+)-3.24 Before undertaking the degradative cleavage of the
side chains, we prepared authentic derivatives containing the
N-acylalanine moiety in 2. Acylation of D- and L-alanine methyl
ester hydrochlorides with the symmetrical anhydride25 of
isovaleric acid furnished (-)-34 and (+)-34 (Scheme 8).
Unfortunately, the specific rotations of these enantiomers were
quite low, and we were reluctant to pursue a degradation strategy
that would rely solely upon this analytical technique. Instead,
we chose to employ the chiral auxiliary (S)-R-methylbenzy-
of the aqueous phase afforded carboxylic acid 37, which without
purification was subjected to diphenylphosphoryl azide (DPPA)-
mediated coupling26,27with (S)-MBA. The formation of amide
(+)-35 established the D configuration of the alanine residue in
(+)-2.
For the synthesis of the side-chain moities of (+)-trienomycin
C, only the S enantiomer of 2-methylbutyric acid was available.
Further exploiting the successful strategy devised for 2, the D-
and L-alanine methyl ester hydrochlorides were first acylated
with the symmetrical anhydride25 of (S)-2-methylbutyric acid
(Scheme 10). Coupling of the resultant methyl esters [(+)-38
and (+)-39] with the dimethylaluminum amide13 of (S)-MBA
then afforded (+)-40 and (-)-41, respectively.
Scheme 8
Scheme 10
lamine (MBA), anticipating that the resultant diastereomers
could be differentiated either spectroscopically or chromato-
graphically. In addition, this group could be readily incorporated
into the requisite authentic samples by acylation of the derived
dimethylaluminum amide13 with the alanine methyl esters (-)-
and (+)-34, providing amides (+)-35 and (+)-36, respectively
(Scheme 8).
With these materials in hand, we attempted to liberate the
intact side chain of (+)-trienomycin B (2). Following saponi-
fication of (+)-2 with KOH (Scheme 9), the organic phase
By analogy with the degradation of 2, saponification of (+)-
trienomycin C (3) with KOH generated (+)-trienomycinol (14),
again identical with the macrocycle derived from (+)-1, as well
as the crude acid 42 (Scheme 11). Coupling of 42 with (S)-
MBA (DPPA)26,27 afforded amide (+)-40, revealing both the D
configuration of the side-chain alanine moiety and the S
configuration at C(30).
Scheme 9
Scheme 11
Isolation and Characterization of a New Congener, (+)-
Trienomycin F. In the course of our stereochemical investiga-
contained (+)-trienomycinol (14), identical in all respects with
the compound obtained via degradation of (+)-1. Acidification

Stereochemistry of Trienomycins and Mycotrienins
J. Am. Chem. Soc., Vol. 118, No. 35, 1996 8313
tions, we discovered that mixtures of (+)-trienomycins B and
C also contained small amounts of a new congener, which we
designated (+)-trienomycin F (13). After HPLC separation,
Scheme 12
1
comparison with 1-3 via 500-MHz H NMR suggested that
13 contained a different acyl fragment attached to the D-alanine
residue. Three distinct resonances were attributable to this
moiety: a quartet of doublets (δ 6.53, J ) 1.4 and 6.9 Hz, 1
H), an apparent triplet (δ 1.83, Japp )1.1 Hz, 3 H), and a doublet
of doublets (δ 1.75, J )1.0 and 6.9 Hz, 3 H).28 In conjunction
with homonuclear decoupling experiments, which verified the
6.9-Hz coupling between the resonances at δ 1.75 and 6.53,
these data were indicative of a 2-methyl-2-butenamide unit.
Comparison of the chemical shift for the C(31) vinylic proton
in (+)-13 with the corresponding resonances for methyl tiglate
(43a), methyl angelate (44a), and the derived N-acyl-D-alanine
methyl esters 43b and 44b29 then established the E geometry
of the C(30,31) olefin (Table 3).
1
Table 3. Vinylic H NMR Chemical Shifts for (+)-13 and Four
Model 2-Methyl-2-butenamides
In principle, the selective C(11) acylation could be effected
with either the complete side chain or an N-protected alanine
derivative. The latter approach would require subsequent
deprotection and N-acylation, whereas the former would rely
upon the DPPA method26,27 described earlier to suppress
epimerization of the N-acylamino acid. This issue was resolved
when DPPA failed to promote coupling of (+)-46 with either
47 or 48,33 the side chains of (+)-trienomycins A and F,
respectively.
compound
R
δ (ppm)
43a
44a
43b
44b
OMe
OMe
D-Ala-OMe
D-Ala-OMe
D-Ala-OR
6.97
6.18
6.48
5.57
6.53
(+)-13
Partial Synthesis of (+)-Trienomycins A and F: A Flexible
End-Game Synthetic Strategy. To confirm the structure of
(+)-trienomycin F (13), we designed a partial synthesis from
(+)-trienomycinol (14), available in adequate quantities from
our degradation work. We anticipated that an end-game strategy
for a unified synthetic approach to the trienomycins would also
emerge from this effort.
At this juncture, we turned to the stepwise route, whereby
initial acylation of the C(11) hydroxyl in (+)-46 with the
symmetrical anhydride25 of BOC-D-alanine34 (DMAP, CH2Cl2,
-50 °C) afforded (+)-49 in 50% yield (Scheme 13). Subse-
quent unmasking of the amine with neat TFA35 resulted in
Our first concern a priori was chemoselective functionaliza-
tion of the C(11) hydroxyl. Importantly, Zeeck and co-workers
employed DCC to achieve the selective C(11) acylation of (+)-
mycotrienol I (8).5b Extension of this tactic to our system would
entail protection of the phenolic hydroxyl in acetonide (+)-15
and liberation of the anti diol prior to the critical acylation step.
In the event, the phenol was readily converted to the methyl
ether (MeI, Ag2O, ∆, 31% yield)12c and MEM ether (MEM Cl,
i-Pr2EtN, CH2Cl2, 77%),30 but attempts to remove these groups
met with limited success. Turning next to a silyl protecting
group, reaction of 15 with tert-butyldimethylsilyl (TBS) triflate31
gave the ether in 71% yield (Scheme 12). In this case, the TBS
group could be removed with TBAF32 in CH2Cl2. Methanolysis
of the acetonide (camphorsulfonic acid, MeOH) then provided
diol (+)-46 (95%).
Scheme 13
(24) From ca. 100 mg of a mixture believed to contain only (+)-
trienomycins B and C, we isolated ca. 3 mg of (+)-trienomycin F.
(25) Bodanszky, M.; Bodanszky, A. The Practice of Peptide Synthesis;
Springer-Verlag: New York, 1984; p 101.
(26) The Weinreb aluminum amide couplings were accompanied by
partial (ca. 10-22%) epimerization at C(28).
(27) (a) See ref. 25, p 247. (b) Shioiri, T.; Ninomiya, K.; Yamada, S. J.
Am. Chem. Soc. 1972, 94, 6203.
(28) Chemical shift data are reported in ppm relative to internal TMS.
(29) Nair, M. D.; Adams, R. J. Am. Chem. Soc. 1961, 83, 922.
(30) Corey, E. J.; Gras, J.-L.; Ulrich, P. Tetrahedron Lett. 1976, 809.
(31) Corey, E. J.; Cho, H.; Ru¨cker, C.; Hua, D. H. Tetrahedron Lett.
1981, 22, 3455.
skeletal rearrangement; acylation with cyclohexane carboxylic
acid and benzotriazol-1-yloxytris(dimethylamino) phosphonium
(32) Corey, E. J.; Venkateswarlu, A. J. Am. Chem. Soc. 1972, 94, 6190.

8314 J. Am. Chem. Soc., Vol. 118, No. 35, 1996
Smith et al.
hexafluorophosphate (BOP)36 then furnished a compound
spectroscopically in accord with the structure (+)-50.37
To circumvent the acid-promoted rearrangement, we em-
ployed the base-labile FMOC protecting group. Selective C(11)
acylation of (+)-46 with the symmetrical anhydride25 of FMOC-
D-alanine38 gave (+)-51 in 56% yield (Scheme 14) along with
Scheme 15
Scheme 14
before, followed by acylation with tiglic acid and BOP to give
(+)-53. Cleavage of the TBS group provided semisyn-
thetic (+)-13 in 68% yield, identical with natural (+)-trieno-
mycin F. The 3R, 11S, 12R, 13R, 28R configurations we
elucidated for (+)-trienomycins A-C were conserved in the
new congener.
Stereochemistry of (+)-Mycotrienins I and II. As in the
trienomycin series, the relative and absolute stereochemistries
of the mycotrienins have remained unknown apart from the
D-alanine side-chain configuration. In an effort to extend the
(+)-trienomycin assignments to the (+)-mycotrienins, we
explored the possibility of chemical correlation via selective
C(19) oxidation of (+)-trienomycin A (1) (Scheme 16).
a 28% combined yield of the C(13)-O-acyl and bis-acyl
products. Liberation of the primary amine by exposure to
diethylamine in CH2Cl2 (1:1, 25 °C, 30 min)38 followed directly
by BOP-mediated coupling with cyclohexanecarboxylic acid
afforded (+)-52 (78% yield, two steps). Desilylation with
TBAF32 then furnished (+)-trienomycin A (1), identical with
1
the natural material in all respects [500-MHz H NMR, 125-
Scheme 16
MHz 13C (INEPT) NMR, IR, UV, HRMS, optical rotation, TLC
(three solvent systems), melting point, and mixed melting point].
Having established the viability of our end-game synthetic
strategy, we next confirmed the structure of (+)-trienomycin F
(13) via an analogous sequence (Scheme 15). Specifically, the
FMOC group in (+)-51 was removed with diethylamine as
(33) Acids 47 and 48 were prepared via coupling of D-alanine methyl
ester with the symmetrical anhydrides of cyclohexanecarboxylic and tiglic
acids, respectively, followed by ester hydrolysis.
(34) BOC-D-alanine was purchased from Schweizerhall, Inc. (South
Plainfield, NJ).
(35) See ref 25, p 170.
(36) (a) Castro, B.; Dormoy, J. R.; Evin, G.; Selve, C. Tetrahedron Lett.
1975, 1219. (b) Castro, B.; Dormoy, J. R.; Dourtoglou, B.; Evin, G.; Selve,
C.; Ziegler, J.-C. Synthesis 1976, 751. (c) Castro, B.; Evin, G.; Selve, C.;
Seyer, R. Synthesis 1977, 413.
(37) A plausible pathway for the formation of (+)-50 under acidic
conditions is illustrated below. The propensity of related model systems to
form γ-lactams has also been observed in our laboratory (Wood, J. L.,
unpublished results).
(38) Chang, C. D.; Waki, M.; Ahmad, M.; Meienhofer, J.; Lundell, E.
O.; Haug, J. D. Int. J. Peptide Protein Res. 1980, 15, 59.

Stereochemistry of Trienomycins and Mycotrienins
J. Am. Chem. Soc., Vol. 118, No. 35, 1996 8315
Oxidation of (+)-1 with Fremy’s salt [(KSO3)2NO]39 fur-
nished (+)-mycotrienin I (6) in 3% yield. Despite partial
decomposition upon exposure to silica, the identity of semi-
synthetic 6 was unambiguously established by comparison [500-
MHz 1H and 125-MHz 13C (INEPT) NMR, IR, HRMS, optical
rotation, and TLC (three solvent systems)] with a sample of
natural (+)-mycotrienin I.40 A second sample of (+)-1 was
likewise treated with Fremy’s salt and then washed with 5%
aqueous sodium dithionite to reduce the resultant quinone to
the hydroquinone. Semisynthetic (+)-mycotrienin II (7), iso-
lated in 13% yield, proved indistinguishable from natural
material as well. These results revealed that the (+)-myco-
trienins share the chiral architecture of the (+)-trienomycins,
differing only in the oxidation state of the aromatic nucleus.
Summary. The complete relative and absolute stereochem-
istries of (+)-trienomycins A, B, and C (1-3) have now been
elucidated. These investigations also led to the discovery and
characterization of a new congener, (+)-trienomycin F (13). The
partial synthesis of (+)-1 and (+)-13 established the viability
of (+)-trienomycinol (14) as an advanced intermediate in a
unified synthetic approach to the trienomycins. In addition, the
trienomycin stereochemical assignments were extended to (+)-
mycotrienin I and II (6 and 7) and hence to (+)-mycotrienols
I and II (8 and 9).41
Acknowledgment. We are pleased to acknowledge financial
support provided by the National Institutes of Health (National
Cancer Institute) through Grant CA-19033. In addition, we
thank Drs. George T. Furst and Patrick J. Carroll and Mr. John
Dykins of the University of Pennsylvania Spectroscopic Service
Center for assistance in securing and interpreting high-field
NMR spectra, X-ray crystal structures, and mass spectra,
respectively.
Supporting Information Available: Complete experimental
details for compounds 1, 6, 7, 13, 15, 16, 19, 23-26a, 26b,
27-36, 38-41, 43b, 44b, 45, 46, 49, and 50-53 (21 pages).
See any current masthead page for ordering and Internet access
instructions.
JA961400I
(39) Zimmer, H.; Lankin, D. C.; Horgan, S. W. Chem. ReV. 1971, 71,
229.
(40) We gratefully acknowledge a generous gift of (+)-mycotrienins I
and II from Professor H. Seto (University of Tokyo).
(41) Mycotrienin II has been converted to mycotrienols I and II: (a) ref
4c. (b) Sugita, M.; Hiramoto, S.; Andoj, C.; Sasaki, T.; Furihata, K.; Seto,
H.; Ohtake, N. J. Antibiot. 1985, 38, 799.