Total synthesis of kendomycin featuring intramolecular doetz benzannulation

Article abstract of DOI:10.1021/ol100229f

One-step formation of the ansa-skeleton realized the synthesis of kendomycin, an ansa-type quinone methide. The Fischer carbene complex derived from the ansa-chain portion was subjected to the intramolecular Doetz benzannulation to afford the desired oxametacyclophane with exclusive regioselectivity. Subsequent Claisen rearrangement, ortho oxidation of the resulting phenol derivative, and mild transformation from p-quinone to p-quinone methide on a silica gel plate furnished kendomycin.

Full text of DOI:10.1021/ol100229f

ORGANIC
LETTERS
2010
Vol. 12, No. 8
1700-1703
Total Synthesis of Kendomycin
Featuring Intramolecular Do¨tz
Benzannulation
Kyosuke Tanaka, Masahito Watanabe, Kodai Ishibashi, Hiroshi Matsuyama,
Yoko Saikawa,* and Masaya Nakata*
Department of Applied Chemistry, Faculty of Science and Technology, Keio
UniVersity, 3-14-1 Hiyoshi, Kohoku-ku, Yokohama 223-8522, Japan
saikawa@applc.keio.ac.jp; msynktxa@applc.keio.ac.jp
Received January 28, 2010
ABSTRACT
One-step formation of the ansa-skeleton realized the synthesis of kendomycin, an ansa-type quinone methide. The Fischer carbene complex
derived from the ansa-chain portion was subjected to the intramolecular Do¨tz benzannulation to afford the desired oxametacyclophane with
exclusive regioselectivity. Subsequent Claisen rearrangement, ortho oxidation of the resulting phenol derivative, and mild transformation from
p-quinone to p-quinone methide on a silica gel plate furnished kendomycin.
Natural ansamycins are quite attractive for organic chemists
and biologists from the point of both their fascinating
structures and impressive pharmacological profiles.1 The
complex ansamycin, kendomycin (1), has a unique quinone
methide architecture connected to a highly substituted
tetrahydropyran and possesses potent antibacterial and cy-
totoxic activities as well as activities as an endothelin receptor
antagonist and an antiosteoporotic agent.2 Since kendomycin
(1) was isolated from Streptomyces species,2 it has been
targeted by a number of synthetic organic chemists, and four
total syntheses3 and one formal total synthesis4 manipulating
each elaborated synthetic strategy have been reported. The
most common route for total syntheses of kendomycin and
other ansamycins generally involves elongation of an ali-
phatic ansa-chain from one or both sides of an aromatic core
followed by macrocyclization. In contrast, we conceived a
unique strategy to construct an ansa-framework featuring an
intramolecular Do¨tz benzannulation5,6 with simultaneous
macrocyclization. This challenging one-step formation of an
ansa-compound from a Fischer-type chromium-carbene
(3) (a) Yuan, Y.; Men, H.; Lee, C. J. Am. Chem. Soc. 2004, 126, 14720–
14721. (b) Smith, A. B., III.; Mesaros, E. F.; Meyer, E. A. J. Am. Chem.
Soc. 2005, 127, 6948–6949. (c) Smith, A. B., III.; Mesaros, E. F.; Meyer,
E. A. J. Am. Chem. Soc. 2006, 128, 5292–5299. (d) Lowe, J. T.; Panek,
J. S. Org. Lett. 2008, 10, 3813–3816. (e) Magauer, T.; Martin, H. J.; Mulzer,
J. Angew. Chem., Int. Ed. 2009, 48, 6032–6036. (f) Magauer, T.; Martin,
H. J.; Mulzer, J. Chem.sEur. J. 2010, 16, 507–519.
(1) For recent reviews on ansamycins, see: (a) Funayama, S.; Cordell,
G. A. In Studies in Natural Products Chemistry; Atta-ur-Rahman, Ed.;
Elsevier Science: Amsterdam, 2000; Vol. 23, pp 51-106. (b) Floss, H. G.;
Yu, T.-W. Chem. ReV. 2005, 105, 621–632. (c) Bryskier, A. In Antimicrobial
Agents; BryskierA., Ed; ASM Press: Washington, DC, 2005; pp 906-
924.
(2) (a) Funahashi, Y.; Kawamura, N.; Ishimaru, T. Jpn. Kokai Tokkyo
Koho 1996, 08231551; Chem. Abstr. 1996, 125, 326518. (b) Funahashi,
Y.; Ishimaru, T.; Kawamura, N. Jpn. Kokai Tokkyo Koho 1996, 08231552;
Chem. Abstr. 1997, 126, 6553. (c) Su, M. H.; Hosken, M. I.; Hotovec, B. J.;
Johnston, T. L. U. S. Patent 5728727, 1998;Chem. Abstr. 1998, 128, 239489.
(d) Bode, H. B.; Zeeck, A. J. Chem. Soc., Perkin Trans. 1 2000, 323–328.
(4) Bahnck, K. B.; Rychnovsky, S. D. J. Am. Chem. Soc. 2008, 130,
13177–13181.
(5) (a) Do¨tz, K. H. Angew. Chem., Int. Ed. Engl. 1975, 14, 644–645.
(b) Waters, M. L.; Wulff, W. D. In Organic Reactions; Overman, L. E.,
Ed.; Jon Wiley & Sons, Inc.: Hoboken, 2008; pp 121-623.
(6) For a partial synthesis of kendomycin using the intermolecular Do¨tz
reaction, see: White, J. D.; Smits, H. Org. Lett. 2005, 7, 235–238
.
10.1021/ol100229f  2010 American Chemical Society
Published on Web 03/12/2010

complex having a long ω-alkynyloxy chain has realized total
synthesis of kendomycin (1), which is the subject of this
communication.
Scheme 2. Synthesis of Left-Half Segment 4
Scheme 1. Synthetic Strategy of Kendomycin (1)
hydrozirconation-iodination4,10 and desilylation gave vinyl
iodide 10. Oxidation to the aldehyde followed by addition
of alkynyllithium afforded the left-half segment 4.
On the other hand, (-)-6 was deoxygenated in two steps,
and the resulting olefin 1111 was transformed into 13 through
a conventional five-step sequence involving Wittig olefination
of ketone 12 (Scheme 3). After iodination of 13, the
protecting group was exchanged to give the right-half
segment 5.
Our synthetic strategy is outlined in Scheme 1. Regarding
kendomycin (1) as an oxametacyclophane (bold lines), we
designed the precursor 2 with an allyl ether positioned as a
foothold for Claisen rearrangement to connect C20 to the
aromatic ring (C20a). The oxametacyclophane 2 was ex-
pected to be the favored product of the intramolecular Do¨tz
benzannulation of chromium-carbene complex 3, whose
alkyne terminus C20a would regioselectively insert to the
carbene center C15b and then the resulting complex would
cyclize via CO insertion between C3 and C4a to give an
aromatic core. Indeed, our previous fundamental studies
using simple substrates demonstrated that vinylidene
chromium-carbene complexes having a sufficient length of
ω-alkynyloxy chain tended to macrocyclize to give oxam-
etacyclophanes while short chains produced oxa-orthocy-
clophanes.7 The excellent regioselectivity of the long tethered
carbene complex prompted us to challenge a highly substi-
tuted complex 3. The complex 3 would be synthesized via
Suzuki-Miyaura coupling between the left- and right-half
segments 4 and 5 followed by chromium complexation.
Scheme 3. Synthesis of Right-Half Segment 5
The Suzuki-Miyaura coupling was realized by palladium-
catalyzed cross-coupling of 4 with the alkyl borane derived
from 5 to cleanly afford the adduct in 98% yield (Scheme
4).3a,4 This was oxidized to ynone 14 from which the
tetrahydropyran 15 was derived by acetal exchange followed
by deoxygenation of the resulting methyl acetal with Et3SiH.
After silylation and selective deprotection, allyl alcohol 16
was introduced to the acetoxychromium carbene complex
prepared from chromate 1712 in situ, giving the key carbene
complex 3 in 95% yield. The crucial Do¨tz reaction was
conducted under our optimized conditions7 (in toluene,
The left- and right-half segments 4 and 5 were synthesized
from (+)- and (-)-enantiomers of the known homoallyl
alcohol 6,8 respectively (Schemes 2 and 3). The terminal
olefin of (+)-6 was oxidatively cleaved after silylation of
the hydroxy group, and the resulting aldehyde was treated
with alkynyllithium prepared from 79 and n-BuLi to give
adduct 8 with a 3.5:1 diastereoselectivity (Scheme 2).
Sequentialacetalization,diimidehydrogenation,andDess-Martin
oxidation led to aldehyde 9. Alkynylation of 9 followed by
(7) Watanabe, M.; Tanaka, K.; Saikawa, Y.; Nakata, M. Tetrahedron
Lett. 2007, 48, 203–206.
(8) (a) Roush, W. R.; Palkowitz, A. D.; Ando, K. J. Am. Chem. Soc.
1990, 112, 6348–6359, for (+)-6. (b) Roush, W. R.; Coffey, D. S.; Madar,
D. J. J. Am. Chem. Soc. 1997, 119, 11331–11332, for (-)-6.
(9) Zeng, F.; Negishi, E. Org. Lett. 2002, 4, 703–706, and references
therein.
(10) Hart, D. W.; Schwartz, J. J. Am. Chem. Soc. 1974, 96, 8115–8116.
(11) Yokokawa, F.; Asano, T.; Okino, T.; Gerwick, W. H.; Shioiri, T.
Tetrahedron 2004, 60, 6859–6880.
(12) Chamberlin, S.; Wulff, W. D.; Bax, B. Tetrahedron 1993, 49, 5531–
5547.
Org. Lett., Vol. 12, No. 8, 2010
1701

Scheme 4
.
Intramolecular Do¨tz Benzannulation and Macrocyclic
Claisen Rearrangement
Scheme 5
.
Quinone Methide Formation and Completion of
Total Synthesis of Kendomycin (1)
task was realized by IBX oxidation15 in EtOH, accompanied
by deprotection of the MOM ether, to afford 2-hydroxy-1,4-
quinone 24 in 82% yield. Furthermore, the subsequent
quinone methide formation also succeeded under the mild
conditions, just applying 24 on silica gel TLC to give 25 in
78% yield (22% recovered 24). Desilylation of 25 furnished
the total synthesis of kendomycin (1) in 91% yield, whose
data were identical with those of the natural sample.
50 °C) under degassed atmosphere to provide the desired
oxametacyclophane 2 in 58% yield as the sole isolable
product. After silylation of 2, the macrocyclic Claisen
rearrangement in the presence of Ac2O and DMAP provided
acetate 18 in 85% yield. Without addition of Ac2O and
DMAP, the unexpected dihydrobenzofuran 19 was obtained
as a major product.13 Aiming at the further conversion, the
resulting 18 was transformed to MOM ether 20.
Next, we addressed the troublesome oxidative cleavage
of the exomethylene group in 20 to the carbonyl group
(Scheme 5). Indeed, the electron-rich, inner double bond was
selectively dihydroxylated to give the diol under all condi-
tions attempted. The diol function was used to mask the inner
double bond, and treatment with O3 yielded the desired
ketone 21. The double bond was then regenerated, and the
resulting 22 was converted to phenol 23 in order to prepare
for the next oxidation.
The climax of the final stage was adjustment of the
oxidation state of the aromatic ring and quinone methide
formation. The final two-step conversion by other research
groups3,4 consisted of a mild oxidation of the catechol
monomethyl ether14 to the unstable o-quinone by IBX or
Dess-Martin periodinane and subsequent acid-catalyzed
quinone methide formation with HF. Our oxidation, in
contrast, required direct ortho oxidation of phenol 23. This
We have a great interest in the last smooth transformation
from p-quinone 24 to p-quinone methide 25 on silica gel
TLC and, therefore, investigated the role of silica gel. After
24 (λmax 401 nm in CHCl3) was applied on silica gel powder,
its yellow color immediately turned red as the solvent
evaporated. We could not isolate the red compound which
easily returns to the yellow 24 by elution with CHCl3 from
silica gel. To clarify the structure of the red intermediate,
the behavior of 24 on silica gel was monitored by UV-vis
spectra on diffuse reflectance mode (Figure 1). The absorp-
tion maximum of the adsorbed red compound was observed
at 528 nm and this absorption gradually decreased as the
pale-yellow quinone methide 25 (λmax 365 nm (sh) in CHCl3)
formed. The resulting mixture was a ca. 3.5:1 ratio of 25
and 24. Since applying the isolated 25 on silica gel powder
also gave the same mixture with the same ratio, 24 and 25
were found to be in equilibrium. We next synthesized simple
two model compounds 26 and 27 to compare their UV
spectra with that of the red intermediate. 2-Hydroxy-1,4-
quinone 2616 (λmax 405 nm in CHCl3) also showed λmax 534
nm on silica gel powder, suggesting that neither the acetal
(15) For formation of o-quinone from monoprotected hydroquinone, see:
Magdziak, D.; Rodriquez, A. A.; Van De Water, R. W.; Pettus, T. R. R.
Org. Lett. 2002, 4, 285–288.
(16) For the synthesis of 26, triethylamine salt of 26, and 27, see the
Supporting Information.
(13) Shih, T. L.; Wyvratt, M. J.; Mrozik, H. J. Org. Chem. 1987, 52,
2029–2033.
(17) The spontaneous irreversible solvolysis of kendomycin (1) in
DMSO-H2O was reported. See Janssen, C. O.; Lim, S.; Lo, E. P.; Wan,
K. F.; Yu, V. C.; Lee, M. A.; Ng, S. B.; Everett, M. J.; Buss, A. D.; Lane,
D. P.; Boyce, R. S. Bioorg. Med. Chem. Lett. 2008, 18, 5771–5773.
(14) Catechol monomethyl ether as an early-stage intermediate of
kendomycin synthesis was prepared via the Do¨tz reaction, see ref 6.
1702
Org. Lett., Vol. 12, No. 8, 2010

nm in CHCl3) rather than that of the MOM-protected
o-quinone 2716 (λmax 465 nm in CHCl3 and λmax 494 nm on
silica gel powder). Based on these data, we considered that
the red intermediate formed as the result of coordination with
silica gel has a charge-delocalized structure similar to
diketonate. The equilibrium between 24 and 25 has not been
observed during preservation of these compounds.17 The
silica gel-promoted charge delocalization would lead to the
smooth equilibrium between 24 and 25 via the acetal
formation/cleavage and tautomerization.
In summary, the total synthesis of kendomycin (1) was
accomplished featuring the intramolecular Do¨tz benzannu-
lation of an unprecedented chromium-carbene complex
tethered to a highly substituted long ω-alkynyloxy chain,
Claisen rearrangement of the macrocycle, ortho oxidation
of phenol, and a mild conversion from p-quinone to
p-quinone methide on silica gel. The success of this
macrocyclization of a highly functionalized carbene complex
demonstrates that this method would be applicable to the
divergent synthesis of ansa-type compounds.
Acknowledgment. We thank Takeda Pharmaceutical Co.
Ltd. for providing us natural kendomycin (TAN 2162) and
Drs. Hiroaki Imai and Yuya Oaki (Keio University) for the
use of the diffuse reflectance UV-vis spectrometer.
Figure 1. Time-course UV-vis spectra of the transformation from
24 to 25 (time after applying 24 on silica gel) and UV-vis
absorption maxima of 26, Et3N salt of 26, and 27.
Supporting Information Available: Experimental pro-
cedures and characterization data for all new compounds.
This material is available free of charge via the Internet at
http://pubs.acs.org.
formation nor the enolization from C20 into any side (C4 or
C19) is essential for the initial color change of 24 (from
yellow to red). Furthermore, the red absorption maximum
was close to that of the triethylamine salt of 2616 (λmax 531
OL100229F
Org. Lett., Vol. 12, No. 8, 2010
1703