Synthesis and structure of tolyporphin A O,O-diacetate.

Article abstract of DOI:10.1021/ol9902374

[formula: see text] The revised structure of tolyporphin A O,O-diacetate (2b) was synthesized by assembling fragments 4, 5, and 12. The synthetic substance was found to be identical to the O,O-diacetate derived from natural tolyporphin A in every respect,

Full text of DOI:10.1021/ol9902374

ORGANIC
LETTERS
1999
Vol. 1, No. 7
1129-1132
Synthesis and Structure of
Tolyporphin A O,O-Diacetate
Wengui Wang and Yoshito Kishi*
Department of Chemistry and Chemical Biology, HarVard UniVersity,
12 Oxford Street, Cambridge, Massachusetts 02138
kishi@chemistry.harVard.edu
Received August 18, 1999
ABSTRACT
The revised structure of tolyporphin A O,O-diacetate (2b) was synthesized by assembling fragments 4, 5, and 12. The synthetic substance was
found to be identical to the O,O-diacetate derived from natural tolyporphin A in every respect, thus establishing the relative and absolute
configurations of this natural product.
In 1992, Moore and co-workers reported the isolation of
tolyporphin A from the lipophilic extract of the cyanophyte
microalga Tolypothrix nodosa. This structurally unique
porphyrin was found to reverse multidrug resistance (MDR)1
in a vinblastine-resistant population of human ovarian
adenocarcinoma cells.2a Subsequently, 10 additional toly-
porphins (B-K) were isolated and were found to possess
varying degrees of anti-MDR activity.2b-d On the basis of
extensive spectroscopic studies, the structure of tolyporphin
A, the representative member of the tolyporphin class of
natural products, was concluded to be 1a.2a We previously
described a total synthesis of the proposed structure of (+)-
tolyporphin A O,O-diacetate (1b) but found that the synthetic
material was not identical to the O,O-diacetate derived from
Subsequently, we carried out extensive spectroscopic
natural (+)-tolyporphin A.3
studies on both the synthetic material and the O,O-diacetate
derived from natural tolyporphin A and proposed the revised
(1) For recent reviews on MDR, see: (a) Pastan, I.; Gottesman, M. M.
structure 2a for tolyporphin A.4 In this Letter, we report a
Annu. ReV. Biochem. 1993, 62, 385. (b) Simon, S. M.; Schindler, M. Proc.
total synthesis of the O,O-diacetate 2b of the revised structure
of tolyporphin A, unambiguously establishing the relative
and absolute configurations of this natural product.5
Natl. Acad. Sci. U.S.A. 1994, 91, 3497.
(2) (a) Prinsep, M. R.; Caplan, F. R.; Moore, R. E.; Patterson, G. M. L.;
Smith, C. D. J. Am. Chem. Soc. 1992, 114, 385. (b) Prinsep, M. R.;
Patterson, G. M. L.; Larsen, L. K.; Smith, C. D. Tetrahedron 1995, 51,
10523. (c) Prinsep, M. R.; Patterson, G. M. L.; Larsen, L. K.; Smith, C. D.
J. Nat. Prod. 1998, 61, 1133. (d) Morliere, P.; Maziere, J.-C.; Santus, R.;
Smith, C. D.; Prinsep, M. R.; Stobbe, C. C.; Fenning, M. C.; Golberg, J.
L.; Chapman, J. D. Cancer Res. 1998, 58, 3571.
(3) Minehan, T. G.; Kishi, Y. Angew. Chem. Int. Ed. 1999, 38, 923.
(4) Minehan, T. G.; Cook-Blumberg, L.; Kishi, Y.; Prinsep, M. R.;
Moore, R. E. Angew. Chem. Int. Ed. 1999, 38, 926.
10.1021/ol9902374 CCC: $18.00 © 1999 American Chemical Society
Published on Web 09/09/1999

Structurally, tolyporphin A consists of an unsymmetrical
dioxobacteriochlorin core with quaternary centers at C.7 and
C.17 containing â-linked C-glycosides. We were interested
in assembling the tolyporphin skeleton from monocyclic
precursors (Figure 1). A tactical advantage of this approach
promoted addition of sterically hindered silyl ketene acetals
to carbohydrate 1-O-acetates and (2) an additional C.9
stereogenic center in the nucleophile can influence the facial
selectivity of the C-glycosidation, thereby installing the
desired quarternary stereogenic center in a highly controlled
manner. Although optimal stereoselectivity and chemical
yield were obtained for the case of 6b + 7b f 8c, synthetic
efficiency was most effectively secured via C-glycoside 8a
since utilization of this precursor mininized the number of
synthetic operations required to access 3. On the basis of
these considerations, we planned the synthesis of the revised
ring-A (and ring-C) precursor 12 via the C-glycosidation of
6a with 9, the antipode of 7b (Scheme 2).
Scheme 2
Figure 1.
is that the ring-C precursor is identical to the ring-A
precursor. The feasibility of this approach was first demon-
strated by the synthesis of a model of the tolyporphin A
chromophore6 and was then extended to the synthesis of the
proposed structure of (+)-tolyporphin A O,O-diacetate (1b).3
To investigate the proposed revised structure of tolyporphin
A, we planned to synthesize the O,O-diacetate 2b in an
analogous fashion to the synthesis of 1b. This strategy would
therefore require ring-A (and the ring-C) precursor 12, the
C.7 stereoisomer of the previously used ring-A fragment 3.
Previously, 3 was prepared via the C-glycosidation of 3,6-
dideoxy-D-galactopyranose (Scheme 1),3,7 a process in which
C-Glycosidation of the acetate 6a7,8 with (S)-silyl ketene
acetal 99 in the presence of TMSOTf and 4 Å molecular
sieves (powder) in a 10:1 mixture of CH2Cl2 and THF at 0
°C yielded the desired C-glycoside 10 as the major product.
This transformation was observed to proceed with 3:1
stereoselectivity at C.1′ and 10:1 stereoselectivity at C.7.
These stereoselectivities were virtually identical to those
observed for the case of 6a + 7a f 8a. Desilylation,
followed by removal of the undesired minor stereoisomers
by chromatography, afforded pure crystalline C-glycoside
11 in 60% overall yield from 6a.8 The C-glycoside 11 was
then converted to the ring-A precursor 12 in seven steps3 in
47% overall yield.
Scheme 1
The stereochemical assignment of 11, and consequently
of 12, was initially made through NMR analysis: (1) the
two new stereogenic centers were selectively introduced.
Related to this process, several previous observations should
be noted: (1) â-C-glycosides can be obtained by Lewis-acid-
(8) The crude acetate 6a, prepared from the corresponding methyl
R-glycoside by treatment with Ac2O-HOAc (3:1) and catalytic H2SO4 at
0 °C was used for this C-glycosidation. The 60% yield of 11 was based on
the methyl R-glycoside used.
(9) (S)-Silyl ketene acetal was prepared from L-(S)-glutamic acid in five
steps by using the known procedure for the (R)-series, see: Ravid, U.;
Silverstein, R. M.; Smith, L. R. Tetrahedron 1978, 34, 1449 and ref 3.
(5) The numbering used in ref 2a is adopted in this paper.
(6) Minehan, T. G.; Kishi, Y. Tetrahedron Lett. 1997, 38. 6811.
(7) Minehan, T. G.; Kishi, Y. Tetrahedron Lett. 1997, 38. 6815.
1130
Org. Lett., Vol. 1, No. 7, 1999

vicinal spin coupling constant J1′,2′ was determined to be 9.2
Hz for 11 and 1.3 Hz for the C.1′ stereoisomer of 11, thereby
establishing the C.1′ stereochemistry, and (2) the H NMR
Scheme 4
1
spectra showed the thiolactam 12 obtained from 11 to be
the stereoisomer of thiolactam 3, thereby concluding the C.7
stereochemistry. Nevertheless, as the stereochemical assign-
ment of tolyporphin A ultimately depended on the stereo-
chemistry installed in the C-glycosidation, the structure of
11 was conclusively determined via X-ray analysis.
Following the protocols previously established, the ring-A
(and -C), -B, and -D precursors 12, 4,6 and 56 were assembled
to yield the precorphin-metal complex 17 (Scheme 3) and
Scheme 3
The synthesis of tolyporphin A O,O-diacetate was com-
pleted via debenzylation of 20, followed by acetylation and
double allylic oxidation, to furnish 2b as a dark-purple solid.
This double allylic oxidation was first employed in the
synthesis of the chromophore model of toylporphin A,6 but
its application to the synthesis of 1b met with substantial
difficulty. This problem was eventually overcome via a two-
then the bacteriochlorin tetrabenzyl ether 20 (Scheme 4). It
should be noted that the synthesis summarized in Schemes
3 and 410 was carried out with a mixture of diastereomers
due to the stereogenic centers at C.3, C.6, C.12, and/or C.16
up to the bacteriochlorin 20, which was the first stereochemi-
cally homogeneous compound after 11. The bright-green
bacteriochlorin 20 was isolated by preparative TLC [silica
gel; hexane-ethyl acetate (2:1, buffered with 1% triethyl-
amine)] in a comparable overall yield from 11 as in the
previous synthesis of the initially proposed structure of
tolyporphin A.3
Figure 2. 1H-NMR spectra (500 MHz, C6D6) of natural and
synthetic tolyporphin A O,O-diacetates.
(10) The product at this stage was a mixture of diastereomers due to the
chiral centers at C.3, C.6, C.12, and/or C.16.
Org. Lett., Vol. 1, No. 7, 1999
1131

[TLC, HR-MS, UV, 1H NMR (Figure 2)] showed the
synthetic O,O-diacetate 2b to be identical to the O,O-
diacetate derived from natural tolyporphin A,12 unambigu-
ously establishing the stereochemistry of tolyporphin A as
the revised structure 2a. In addition, the CD spectra (Figure
3) demonstrated that the synthetic O,O-diacetate 2b possessed
the same absolute configuration as the O,O-diacetate derived
from natural tolyporphin A. Since the acetate 6a was
prepared from 3,6-dideoxy-D-galactopyranose,6 the absolute
configuration of tolyporphin A is represented by the structure
2a. Strictly speaking, the structural conclusions from the
current work should be limited to tolyporphin A, but we
speculate these conclusions can be applied to other members
of this class of natural products as well.
Acknowledgment. We thank Mr. G. Topalov for the
X-ray analysis of 11 and Dr. J. Guo in the Nakanishi group
at Columbia University for the measurements of CD spectra
of 2b. Financial support from the National Institutes of Health
(CA 22215) is gratefully acknowledged.
Figure 3. CD spectra (CH2Cl2) of natural and synthetic tolyporphin
A O,O-diacetates.
step allylic oxidation.3 Interestingly, it was found that
substrate 21 behaved toward the oxidant in a fashion
analogous to that of the chromophore model, to give synthetic
tolyporphin O,O-diacetate (2b) as a dark-purple solid in a
single step.11
Supporting Information Available: Experimental pro-
cedures summarized in Scheme 2, 3, and 4, spectroscopic
data, and an ORTEP drawing of 11. This material is available
free of charge via the Internet at http://pubs.acs.org.
OL9902374
Extensive chromatographic and spectroscopic studies
(12) We gratefully thank Dr. Miche`le Prinsep for kindly providing a
sample of natural tolyporphin A. Acetylation of tolyporphin A is known to
form tolyporphin A O,O-diacetate (2b).
(11) TLC analysis showed that the oxidation was very clean but the
isolated yield was only around 30%.
1132
Org. Lett., Vol. 1, No. 7, 1999