ATPase activity induced by goniodomin A appears to be
highly sensitive to the troponin/tropomyosin complex.4
Furthermore, goniodomin A has been demonstrated to induce
largely different modulation of actomyosin ATPase activity
between ventricular and atrial muscles.5 It has also been
reported that goniodomin A causes morphological changes
in human astrocytoma cells by increasing the filamentous
actin content.6 In 2002, goniodomin A was reported to inhibit
angiogenesis by the inhibition of endothelial cell migration
and basic fibroblast growth factor-induced tube formation
and to be active in vivo.7 These results clearly indicated that
goniodomin A has profound effects on reorganization of the
cytoskeleton.
The important biological properties of this compound as
well as its intriguing molecular structure prompted us to
embark on the complete stereochemical assignment, includ-
ing the absolute configuration and total synthesis of go-
niodomin A. Recently, Fujiwara and co-workers reported
their synthetic approach to the stereochemical assignment
of goniodomin A.8 Herein, we describe the complete stere-
ochemical assignment of goniodomin A, including the
absolute configuration, based on detailed 2D NMR analysis
of the natural product, synthesis of suitable model compounds
for NMR comparison, degradation experiments, and cor-
relation with synthetic reference compounds.
The relative stereochemistry of the macrolide ring, con-
taining the A- to E-rings, and the isolated hemiacetal F-ring
was determined through analysis of proton coupling constants
and ROESY data of 1 (Figure 2). Due to the conformational
restrictions imposed by the presence of one spiroacetal ring
and three ether rings incorporated into the macrolide ring,
compound 1 appears to be a relatively rigid ring. In particular,
transannular ROESY correlations observed for 1 were keys
to assignment of the relative stereochemistries.
C2-C15 Portion (A). A large coupling constant (J ) 8.4
Hz) between H5 and H6 indicated a diaxial arrangement for
these protons,9 and thus an equatorial orientation of 5-OH.
H2 was also assigned as axial on the basis of ROESY
correlations between H2/H4a and H2/H6. H7 and 9-Me were
assigned as axial by a ROESY cross-peak H7/9-Me. A
ROESY correlation between H10a and 12 ) CHa indicated
the equatorial orientation of the C11-C12 bond, thereby
assigning the acetal oxygen in the C-ring as axial. A ROESY
correlation was observed between H13b10 and H15, suggest-
ing that the C-ring adopts a twist-boat conformation and that
Figure 2. ROESY correlations for the C2-C15 (A), C7-C24 (B),
C24-C31 (C), and C32-C36 (D) portions of goniodomin A (1)
(600 MHz, C6D6, acetone-d6, or CD2Cl2)
the stereochemistry of H15 must have a pseudoaxial orienta-
tion. The relative configuration between the A- and B-rings
was assigned by a large coupling constant (J ) 8.4 Hz)
between H6 and H7, indicating an anti arrangement for these
protons,10 and ROESY correlations H5/H7, H6/8 ) CHa,
and H2/8 ) CHa.
C15-C24 Portion (B). The D-ring was readily assigned
to be a 2,6-cis-substituted dihydropyran by a ROESY
correlation between H16 and H20. The relative stereochem-
istry of the C-/D-ring juncture was assigned on the basis of
a large coupling constant (J ) 9.6 Hz) between H15 and
H16 and ROESY cross-peaks H14a/H16, H14b/H17a, and
H15/H17b. Furthermore, H7 showed significant ROESY
cross-peaks to H16 and H20, allowing us to interrelate the
relative stereochemistries of the BC- and D-rings, as shown.
A large coupling constant (J ) 9.0 Hz) between H20 and
H21 revealed an anti arrangement for these protons, and
ROESY correlations between H20/H22a and H20/H24 al-
lowed the E-ring to be a 2,5-trans-substituted tetrahydrofuran.
In addition, an unobserved ROESY correlation between 19H
and 22-CH2 indicated that C19 and C22 have an anti
arrangement, which was also supported by a transannular
NOE between H7/H22 observed in the ROESY spectrum.
C24-C31 Portion (C). The relative stereochemistry of
the 26,27-diol in 1 was unambiguously determined by its
conversion to the corresponding 26,27-O-isopropylidene
derivative 2 (Scheme 1). A threo relationship of 26,27-diol
was established by a large coupling constant (J ) 8.4 Hz)
between H26 and H27 and observed NOE data of acetonide
2, as shown. The 13C NMR chemical shifts for the dimethyl
groups on the dioxolane ring (27.7 and 27.2 ppm, respec-
tively), which were typical values for a 1,2-threo-diol
system,11 also supported the assigned configuration. Coupling
(3) Furukawa, K.-I.; Sakai, K.; Watanabe, S.; Maruyama, K.; Murakami,
M.; Yamaguchi, K.; Ohizumi, Y. J. Biol. Chem. 1993, 268, 26026-26031.
(4) Yasuda, M.; Nakatani, K.; Matsunaga, K.; Murakami, M.; Momose,
K.; Ohizumi, Y. Eur. J. Pharmacol. 1998, 346, 119-123.
(5) Matsunaga, K.; Nakatani, K.; Murakami, M.; Yamaguchi, K.;
Ohizumi, Y. J. Pharmacol. Exp. Ther. 1999, 291, 1121-1126.
(6) Mizuno, K.; Nakahata, N.; Ito, E.; Murakami, M.; Yamaguchi, K.;
Ohizumi, Y. J. Pharm. Pharmacol. 1998, 50, 645-648.
(7) Abe, M.; Inoue, D.; Matsunaga, K.; Ohizumi, Y.; Ueda, H.; Asano,
T.; Murakami, M.; Sato, Y. J. Cell. Physiol. 2002, 190, 109-116.
(8) (a) Fujiwara, K.; Naka, J.; Katagiri, T.; Sato, D.; Kawai, H.; Suzuki,
T. Bull. Chem. Soc. Jpn. 2007, 80, 1173-1186. (b) Katagiri, T.; Fujiwara,
K.; Kawai, H.; Suzuki, T. Tetrahedron Lett. 2008, 49, 233-237.
(9) Matsumori, N.; Kaneno, D.; Murata, M.; Nakamura, H.; Tachibana,
K. J. Org. Chem. 1999, 64, 866-876.
(10) JH13a, H13b ) 12.6 Hz, JH13a, H14 ) 4.5 Hz, JH13a, H14′ ) 3.6 Hz, JH13b,
H14 ) 10.8 Hz, JH13b, H14′ ) 10.8 Hz (900 MHz, acetone-d6).
(11) Dana, G.; Danechpajouh, H. Bull. Soc. Chim. Fr. 1980, 395-399.
Org. Lett., Vol. 10, No. 5, 2008
1014