Efficient construction of the kedarcidin chromophore ansamacrolide

Article abstract of DOI:10.1021/ol0477374

(Chemical Equation Presented) The streamlined assembly of the ansamacrolide framework of the kedarcidin chromophore via an efficient atropselective Sonogashira coupling step is described. To this end, two newly improved practical syntheses of the cyclopen

Full text of DOI:10.1021/ol0477374

ORGANIC
LETTERS
2005
Vol. 7, No. 2
267-270
Efficient Construction of the Kedarcidin
Chromophore Ansamacrolide
Yasuhito Koyama, Martin J. Lear,*,† Fumihiko Yoshimura, Isao Ohashi,
Tomoko Mashimo, and Masahiro Hirama*
Department of Chemistry, Graduate School of Science, Tohoku UniVersity, COE,
Sendai 980-8578, Japan
martin@ykbsc.chem.tohoku.ac.jp; hirama@ykbsc.chem.tohoku.ac.jp
Received November 4, 2004
ABSTRACT
The streamlined assembly of the ansamacrolide framework of the kedarcidin chromophore via an efficient atropselective Sonogashira coupling
step is described. To this end, two newly improved practical syntheses of the cyclopentene and diyne fragments have been developed, which
feature 0.2 mol % catalytic loadings for an RCM step and i-PrMgCl/CH2I2 as a new entry to gem-disubstituted epoxides from ketones, both
being applicable to 49-g scales.
The kedarcidin chromophore (1) is a highly unstable,
complex natural product that possesses, among other struc-
tural challenges, a conformationally defined ansamacrocyclic
bridge.1 Although 1 has yet to be completed in its entirety,
Myers et al. have constructed the whole aglycon unit of 1
through an impressive transannulation protocol.2 In our work,
we have developed direct R-selective protocols for both of
the 2-deoxysugars3 and have centered on the established
success of the CeCl3/LiN(TMS)2-mediated cyclization pro-
tocol to form the nine-membered, bicyclic core.4 As part of
our intensive efforts to provide gram quantities of late-stage
targets, we describe in this letter the practical preparation of
both the cyclopentene and diyne fragments and the advantage
of having an appropriate C4-stereochemistry5 for the efficient
assembly of the ansamacrolide framework 2.6
In this and our earlier work, the kedarcidin chromophore
1 was retrosynthetically stripped of its sugar components and
reduced to the ansamacrolide subtargets 2 and 3 (Scheme
1). Among several macrolide-forming methods tested, the
intramolecular Sonogashira coupling was the only method
found powerful enough to successfully prepare ansamac-
rolides such as 3. In the earlier cases tested, the C4-epimer
of 3 was targeted and yields fell into the 40-45% range.
Subsequent efforts at improving the coupling yield were
† Current address: Department of Chemistry, 3 Science Drive 3, National
University of Singapore, Republic of Singapore 117543. Email: chmlmj@
nus.edu.sg.
(1) Isolation of kedarcidin and chromophore structure (1): (a) Lam, K.
S.; Hesler, G. A.; Gustavson, D. R.; Crosswell, A. R.; Veitch, J. M.; Forenza,
S.; Tomita, K. J. Antibiot. 1991, 44, 472. Hofstead, S. J.; Matson, J. A.;
Malacko, A. R.; Marquardt, H. J. Antibiot. 1992, 45, 1250. (b) Leet, J. E.;
Schroeder, D. R.; Langley, D. R.; Colson K. L.; Huang, S.; Klohr, S. E.;
Lee, M. S.; Golik, J.; Hofstead, S. J.; Doyle, T. W.; Matson, J. A. J. Am.
Chem. Soc. 1993, 115, 8432. (c) Structure revision and synthesis of
fragments 5 and 7: Kawata, S.; Ashizawa, S.; Hirama, M. J. Am. Chem.
Soc. 1997, 119, 12012.
(2) Myers, A. G.; Hogan, P. C.; Hurd, A. R.; Goldberg, S. D. Angew.
Chem., Int. Ed. 2002, 41, 1062.
(3) (a) L-Mycarose glycosylation study: Lear, M. J.; Yoshimura, F.;
Hirama, M. Angew. Chem., Int. Ed. 2001, 40, 946. (b) L-Kedarosamine
glycosylation study: Ohashi, I.; Lear, M. J.; Yoshimura, F.; Hirama, M.
Org. Lett. 2004, 719.
(4) Examples in advanced synthetic settings: (a) Kawata, S.; Yoshimura,
F.; Irie, J.; Ehara, H.; Hirama, M. Synlett 1997, 250. (b) Kobayashi, S.;
Ashizawa, S.; Takahashi, Y.; Suigura, Y.; Nagaoka, M.; Lear, M. J.; Hirama,
M. J. Am. Chem. Soc. 2001, 123, 11294. (c) Das, P.; Mita, T.; Lear, M. J.;
Hirama, M. Chem. Commun. 2002, 2624. (d) Inoue, M.; Kikuchi, T.;
Hirama, M. Tetrahedron Lett. 2004, 45, 6439. (e) Inoue, M.; Sasaki, T.;
Hatano, S.; Hirama, M. Angew. Chem., Int. Ed. 2004, 43, 6500.
(5) This structural change also arose from unforeseen stereochemical
difficulties that occurred during advanced stages directed toward the
synthesis of the aglycon of 1. This work will be detailed elsewhere.
(6) For our preliminary synthetic study of the C4-epimer of 3, see:
Yoshimura, F.; Kawata, S.; Hirama, M. Tetrahedron Lett. 1999, 40, 8281.
10.1021/ol0477374 CCC: $30.25
© 2005 American Chemical Society
Published on Web 12/30/2004

Scheme 1. Retrosynthesis of Kedarcidin Chromophore 1 to Atropisomeric Ansamacrolide Subtargets 2 and 3
unrewarding. Eventually we surmised that with an ap-
propriately protected precursor the targeting of the C4-epimer
as shown (3) would reduce the ensuing transannular interac-
tions between the chloropyridine and the projecting C8-
hydrogens and likely improve the coupling yield.5
Encouraged by modeling studies on 3,7 we set out on the
development of efficient routes that would enable the large-
scale production of the requisite fragments 4-7 (Scheme
1). The key fragments that mostly required reassessment in
their synthesis were identified as the cyclopentene (4) and
diyne (6) fragments (Scheme 1). Targeting the highly
oxygenated cyclopentene 4, we began by developing a
practical synthesis of the useful building block 13, which
we had previously established as a good precursor to 4
(Scheme 2).6,8
The methyl glycoside 8 was first protected as its known
MPM-acetal10 and the remaining hydroxy groups were
MOM-protected. Chemoselective removal of the MPM-acetal
to the diol 9 and introduction of iodine at the primary position
yielded the iodide 10. Reductive opening of 10 with zinc
dust11 under hydrolytic conditions furnished the alkene-
aldehyde 11 in its hemiacetal form, which was directly
olefinated to efficiently afford the bisalkene 12.
Next, various RCM conditions for 12 were investigated.12
When Grubbs’ first generation catalyst Cl2(PCy3)2RudCHPh
was used, an enone (cf. 20, Scheme 4) was generated as a
side product with 13. Although this result was desirable for
the purpose of making the cyclopentene fragment 4, this
reaction could not be adequately controlled on a large scale.
Having performed several trials, the optimized conditions
to 13 involved using catalytic amounts of Grubbs’ second
generation catalyst 14 in 0.05 M CH2Cl2 at ambient tem-
peratures. After 3 days, purification of the reaction over
activated carbon13 gave excellent yields of the desired
cyclopentenol 13.
Scheme 2. Practical Synthesis of Cyclopentenol 13
The efficiency and practicality of this route to 13 is
noteworthy. Each reaction involves acceptable, non-labor-
intensive purification and workup procedures on multigram
scales. In particular, we have regularly achieved yields of
13 surpassing 90% on 50-g scales by using only 0.2 mol %
of Grubbs’ catalyst 14 in the final RCM step.
(7) By using conformational and minimization searches with MacroModel
8.5 under the MM2* force-field, 3 was estimated to display greater degrees
of freedom, a lower global energy minimum, and a larger transannular cavity
than the C4-macrolide-epimer, epi-3.
(8) For our previous synthesis of chiral trihydroxylated cyclopentenes
akin to 13, see: Toyama, K.; Iguchi, S.; Sakazaki, H.; Oishi, T.; Hirama,
M. Bull. Chem. Soc. Jpn. 2001, 74, 997.
(9) For example: (a) Ovaa, H.; Lastdrager, B.; Codee, J. D. C.; van der
Marel, G. A.; Overkleeft, H. S.; van Boom, J. H. J. Chem. Soc., Perkin
Trans. 1 2002, 2370. (b) Ramana, G. V.; Rao, B. V. Tetrahedron Lett.
2003, 44, 5103. (c) Moon, H. R.; Choi, W. J.; Kim, H. O.; Jeong, L. S.
Chem. Lett. 2004, 506. (d) Nayek, A.; Bauerjee, S.; Sinha, S.; Ghosh, S.
Tetrahedron Lett. 2004, 45, 6457.
(10) (a) Johansson, R.; Samuelsson, B. J. Chem. Soc., Perkin Trans. 1
1984, 2371. (b) Chen, J.; Dorma´n, G.; Prestwich, G. D. J. Org. Chem. 1996,
61, 393.
(11) Ferrier, R. J.; Schmidt, P.; Tyler, P. C. J. Chem. Soc., Perkin Trans.
1 1985, 301.
(12) Selected reviews on metathesis: (a) Schrock, R. R. Tetrahedron
1999, 55, 8141. (b) Furstner, A. Angew. Chem., Int. Ed. 2000, 39, 3012.
(c) Grubbs, R. H. Tetrahedron 2004, 60, 7117.
During the rising popularity of RCM, syntheses of products
akin to 138 became simplified, and several reports detailing
related routes have emerged.9 Independent to these reports,
we devised an efficient route to 13 from the convenient sugar
8 and highlight here the power of RCM to large-scale
synthesis.
(13) Cho, J. H.; Kim, B. M. Org. Lett. 2003, 5, 531.
268
Org. Lett., Vol. 7, No. 2, 2005

To establish a practical, stereoselective synthesis of the
diyne fragment 6, we employed some of our previous
chemistry. In short, the enediyne-diol 16 was reliably
prepared in 40-45% yield in four steps from the readily
available glyceraldehyde derivative 15 (Scheme 3).14 The
enantiomerically pure enediyne diol 16 was subjected to
Sharpless asymmetric epoxidation, (-)-SAE, using diethyl
D-(-)-tartrate at -20 °C. This furnished the â-epoxide 17
in 92% yield over two steps, after selective MMTr protection
of the primary alcohol. Subsequent MOM protection of the
secondary alcohol gave 18. The key step, the regio- and
stereoselective reductive ring opening of the â-epoxide 18,
was found most effective using our first reagent of choice,
namely, Superhydride (LiHBEt3), which under high dilution
conditions in Et2O at low temperature yielded the tertiary
alcohol 19 with only traces of allene side products.15
Deprotection of the MMTr group then afforded the desired
diyne fragment 6 in 88% overall yield.
Scheme 3. Practical Synthesis of Diyne Diol 6
Having synthesized sufficient quantities of the requisite
fragments 4-7,1c,6 the task in hand was to determine whether
our revised plan would result in a more efficient assembly
of the ansamacrolide 3 (Scheme 4). Following TPAP-
oxidization16 of 13 to its enone and introduction of iodine
to give the iodo-enone 20, we needed to produce the
â-epoxide 21 on a gram scale; however, this could not be
achieved reliably under the classic conditions of using the
sulfonium ylid of Me3SI.6,17 After several less productive
routes and reagents were surveyed, we discovered that the
simple generation (under the techniques developed by
Knochel’s group) and addition of R-iodomethylmagnesium
halide to the enone 20 yielded the â-epoxide 21 as a single
isomer in good yields.18 Importantly, this procedure could
be reliably performed starting with gram quantities of 20
(up to 49-g scales have been tested). To the best of our
knowledge, the combination of CH2I2 and i-PrMgCl is new
and represents a convenient entry for the generation of gem-
disubstituted epoxides.18b
Exposure of 21 to aqueous acid afforded the diol 22, which
was protected as its cyclic carbonate, and the MOM groups
were replaced with TBS ethers to furnish 23. Further
successive dual exchanges of diol protective groups and
epoxide formation under Mitsunobu conditions gave 4. This
sequence was found far more productive than that previously
established6 and importantly avoided migration of protecting
groups. Coupling of 4 with the â-2-chloroazatyrosine deriva-
tive 51c according to the previously reported conditions (CsF,
DMF, 60 °C)19 and subsequent TBS protection of the
secondary alcohol gave the aryl ether adduct 24 in 90% yield
over two steps.
Saponification of the methyl ester gave the carboxylic acid
of 24, which was condensed selectively with the primary
alcohol of the diyne 6 using EDC‚HCl and DMAP. Subse-
quent silylation of the tertiary C4-OH with TESCl then gave
the cyclization precursor 25. Although a whole spectrum of
palladium-coupling conditions were screened, the intramo-
lecular Sonogashira coupling of 25 according to the previ-
ously established conditions, Pd2(dba)3‚CHCl3 (0.5 equiv),
CuI (2 equiv), and i-Pr2NEt (30 equiv) in degassed DMF (2
mM) at room temperature, still provided the optimal results
and gave the ansamacrolide 3 as a single atropisomer in 90%
yield. This yield was far superior to that ever obtained for
the same transformation to C4-epi-3 (47% maximum) and
points to reduced transannular interactions.6,7
After several trials, the selective deprotection of the Boc
group of 3 (R) Boc) was efficiently achieved by initial
treatment with TBSOTf to give the incipient O-silyl car-
bamate 3 (R) C(O)OTBS), which was then hydrolytically
collapsed on exposure to silica to give the free amine 3 (R
) H).20 The condensation of this amine 3 (R ) H) with the
naphthoic acid 71c using EDC‚HCl and HOAt21 ultimately
gave the amide 2 in good overall yield.
(14) Wang, G. X.; Iguchi, S.; Hirama, M. J. Org. Chem. 2001, 66, 2146.
(15) (a) For the classical use of LiHBEt3 to open epoxides, see:
Krishnamurthy, S.; Schubert, R. M.; Brown, H. C. J. Am. Chem. Soc. 1973,
95, 6. (b) In THF, moderate yields of 19 were obtained (70-75%)
accompanied by 15-20% of allene stereoisomers due to competing
conjugate additions. Et2O was found to be the optimal solvent.
(16) Ley, S. V.; Norman, J.; Griffith, W. P.; Masden, S. P. Synthesis
1994, 639.
(17) Corey, E. J.; Chaykovsky, M. J. J. Am. Chem. Soc. 1965, 87, 1353.
(18) (a) For recent progress of using i-PrMgCl to generate Grignard
reagents via iodine-magnesium exchange, see: Sapountzis, I.; Lin, W.;
Fischer, M.; Knochel, P. Angew. Chem., Int. Ed. 2004, 43, 4364. (b) For a
related report of using RCHI2/i-PrMgCl to form 1,2-disubstituted epoxides
from aldehydes, see: Schulze, V.; Nell, P. G.; Burton, A.; Hoffmann, R.
W. J. Org. Chem. 2003, 68, 4546. (c) The stereochemical result (R-facial
attack) is similar to that found when sufonium ylids were used and is
presumably dominated by the Burgi-Dunitz trajectory being more hindered
by the C11-MOM ether on the â-face.
The stereochemistry of the atropisomer 2 was unambigu-
1
ously determined by NOE H NMR experiments. In com-
parison to the large NOE values measured between H4′
and H10 in C4-epi-2, smaller NOEs occurred in 2 in the
0.2-0.3% range.6,22 This result again implies a larger
transannular cavity for 2 than the more enclosed cavity of
(19) Kawata, S.; Hirama, M. Tetrahedron Lett. 1998, 39, 8707.
(20) Sakaitani, M.; Ohfune, Y. J. Org. Chem. 1990, 55, 870.
(21) Because of the decomposition of 3 (R ) H), HOAt was found
superior to 1-hydroxybenzotriazole (HOBt): Carpino, L. A. J. Am. Chem.
Soc. 1993, 115, 4397.
Org. Lett., Vol. 7, No. 2, 2005
269

Scheme 4. Assembly of Ansamacrolide 2
its C4-epimer.7 In addition, variable-temperature NMR
Acknowledgment. This work was supported under the
CREST program from JST and a Grant-in-Aid for Scientific
Research, Wakate B, by MEXT (to M.J.L.). Fellowships (to
F.Y. and I.O.) from the Japanese Society for the Promotion
of Science for Young Japanese Scientists are gratefully
acknowledged.
studies were performed on the single atropisomers 2 and 3,
which were found to be conformationally stable about the
azatyrosine stereogenic axis from 25 to 100 °C (see Sup-
porting Information).23
In conclusion, the atropselective construction of the
ansamacrolide moiety of kedarcidin chromophore has been
established in both an efficient and practical manner. Further
studies directed toward the total synthesis of 1 are actively
being pursued and will be disclosed in due course.5
Supporting Information Available: Full experimental
details for all new compounds and reactions are provided,
including 1H NMR spectra of 2 at variable temperatures. This
material is available free of charge via the Internet at
http://pubs.acs.org.
(22) Similarly, between the C4-epimers of 3, only a small (0.3%) NOE
between H4′ and H10 was observed in the epimer 3, unlike the large NOEs
observed for C4-epi-3 (cf. ref 6).
(23) For other observations of atropisomer behaviour, see: Myers, A.
G.; Hurd, A. R.; Hogan, P. C. J. Am. Chem. Soc. 2002, 124, 4583.
OL0477374
270
Org. Lett., Vol. 7, No. 2, 2005