Progress toward synthesis of diazonamide A. Preparation of a 3-(oxazol-5-yl)-4-trifluoromethyl-sulfonyloxyindole and its use in biaryl coupling reactions

Article abstract of DOI:10.1021/ol005548p

The synthesis of a 3-oxazol-5-yl-indole-4-triflate is described, featuring a Schoelkopf reaction to prepare the oxazole. In addition, the participation of this intermediate in biaryl coupling reactions toward the total synthesis of the natural product dia

Full text of DOI:10.1021/ol005548p

ORGANIC
LETTERS
2000
Vol. 2, No. 8
1033-1035
Progress toward Synthesis of
Diazonamide A. Preparation of a
3-(Oxazol-5-yl)-4-trifluoromethyl-
sulfonyloxyindole and Its Use in Biaryl
Coupling Reactions
Edwin Vedejs* and David A. Barda
Department of Chemistry, UniVersity of Wisconsin-Madison, 1101 UniVersity AVenue,
Madison, Wisconsin 53706
edVed@umich.edu
Received January 13, 2000
ABSTRACT
The synthesis of a 3-oxazol-5-yl-indole-4-triflate is described, featuring a Scho1lkopf reaction to prepare the oxazole. In addition, the participation
of this intermediate in biaryl coupling reactions toward the total synthesis of the natural product diazonamide A is presented.
Diazonamide A (1) (Scheme 1) is a highly potent cytotoxic
isolate of the colonial ascidian Diazona chinesis.1 Several
groups including our own have initiated strategies based on
palladium-catalyzed aryl coupling reactions to assemble the
C(16)-C(18) bond of 1.2 In the preceding communication
we described the stereoselective synthesis of 2, an ap-
propriately substituted benzofuranone intermediate possessing
the required C(10) quaternary center and a C(16) halogen
substituent.3 The latter provides the means for eventual
conversion into a reactive organometallic derivative suitable
for coupling to C(18). We now present the preparation of
an indole-oxazole conjugate 3 and demonstrate its ability
to participate in biaryl coupling reactions corresponding to
the formation of the C(16)-C(18) linkage in 1.
Our first approach to a 3-(oxazol-5-yl)indole was based
on the Scho¨llkopf reaction of indole-3-carboxylates with
LiCH2NC. The prospects were tested in a model study using
4 as the substrate.4 Indole nitrogen was protected with the
bulky triisopropylsilyl (TIPS) group because N-TIPS sub-
stitution in pyrroles and indoles is known to prevent
deprotonation at C(2).5 By using this strategy, the addition
of 4 to 5 equiv of LiCH2NC at -78 °C in THF resulted in
an 85% yield of 5 with concomitant cleavage of the
silylamine upon workup.
We then attempted to apply this transformation to the
corresponding 4-iodo indole 6. The latter was prepared by
the directed electrophilic aromatic substitution of 3-formyl-
indole with thallium(III) trifluoroacetate,6 followed by
sodium chlorite oxidation of the formyl group to the acid,
(1) Lindquist, N.; Fenical, W.; Van Duyne, G. D.; Clardy, J. J. Am. Chem.
Soc. 1991, 113, 2303.
(2) (a) Moody, C. J.; Doyle, K J.; Elliott, M. C.; Mowlem, T. J. Pure
Appl. Chem. 1994, 66, 2107. (b) Konopelski, J. P.; Hottenroth, J. M.; Oltra,
H. M.; Veliz, E. A.; Yang, Z.-C. Synlett 1996, 609. (c) Moody, C. J.; Doyle,
K. J.; Elliott, M. C.; Mowlem, T. J. J. Chem. Soc., Perkin Trans. 1 1997,
2413. (d) Boto, A.; Ling, M.; Meek, G.; Pattenden, G. Tetrahedron Lett.
1998, 39, 8167. (e) Wipf, P.; Yokokawa, F. Tetrahedron Lett. 1998, 39,
2223. (f) Hang, H. C.; Drotleff, E.; Elliott, G. I.; Ritsema, T. A.; Konopelski,
J. P. Synthesis 1999, 398. (g) Jeong, S.; Chen, X.; Harran, P. G. J. Org.
Chem. 1999, 63, 8640. (h) Magnus, P.; Kreisberg, J. D. Tetrahedron Lett.
1999, 40, 451. Magnus, P.; McIver, E. G. Tetrahedron Lett. 1200, 41, 831.
Chan, F.; Magnus, P.; McIver, E. G. Tetrahedron Lett. 1200, 41, 835.
(3) Vedejs, E.; Wang, J. Org. Lett. 2000, 2, 1031.
(4) Scho¨llkopf, U.; Schro¨der, R. Angew. Chem., Int. Ed. Engl. 1971, 10,
333.
(5) Pyrroles: (a) Kozikowski, A. P.; Cheng, X. M. J. Org. Chem. 1984,
49, 3239. Indoles: (b) Amat, M.; Sathyanarayana, S.; Hadida, S.; Bosch,
J. Heterocycles 1996, 43, 1713. (c) Iwao, M. Heterocycles 1993, 36, 29
and references therein.
10.1021/ol005548p CCC: $19.00 © 2000 American Chemical Society
Published on Web 03/28/2000

Scheme 1
Scheme 2
conversion to the methyl ester with TMSCHN2, and intro-
duction of the TIPS group. However, attempts to effect the
Scho¨llkopf reaction of 6 using the same conditions as before
resulted in rapid decomposition, and none of the expected
product 7 could be isolated. Evidently, the 4-iodo substituent
impedes the desired addition at the ester carbonyl group.
At this point, attention was turned to the use of a triflate
as the activating group at C(18) (diazonamide numbering).
The triflate approach (Scheme 2) has the advantage that
indoles containing the C(4) oxygen (indole numbering) are
commercially available and can be prepared conveniently
using the Leimgruber-Batcho procedure.7 Benzyloxyindole
8 was chosen as the starting material and was converted to
the TIPS silylamine. Bromination at C(3) with NBS in THF
then occurred efficiently to give 9 (90% overall from 8).8
Subsequent low-temperature lithium-halogen exchange with
t-BuLi, followed by reaction with methyl chloroformate,
resulted in 10 in 82% yield.
The Scho¨llkopf reaction of 10 was carried out using the
same conditions that had been developed for the C(4)-
unsubstituted indole ester 4. In contrast to 4, the 4-benzyl-
oxyindole 10 gave a 2:1 mixture of the keto isocyanide 11
and the desired oxazole 12 (71% combined yield). If the
mixture of 11 and 12 was subjected to mild acid treatment
(pyridinium toluenesulfonate, CH2Cl2), 11 was completely
converted to 12 with no need for isomer separation.
Ultimately, the oxazolyl indole 12 was obtained in 71%
overall yield from 10. Further elaboration to the aryl triflate
3 could then be carried out in three steps. First, the indole
nitrogen was protected using NaH and Boc2O to produce
the N-BOC derivative. The benzyl ether was cleaved (Pd/C,
H2) to the free phenol 13 (quantitative) and treatment with
NaH and (F3CSO2)2NPh (14) gave the triflate 3 in 81% yield
from 12.
To test the reactivity of a 4-trifluoromethylsulfonyloxy-
indole for the desired cross coupling, 3 was submitted to
Suzuki coupling conditions with boronic acid 14.9 The best
results were obtained using Pd(dppf)Cl2/K3PO4 in refluxing
anhydrous THF.10 This procedure gave an excellent 92% yield
of the biaryl product 15. Aqueous solvent combinations were
not suitable for the coupling reaction because of rapid
cleavage of the N-BOC protecting group.
The final series of experiments was designed to test the
feasibility of indole triflate coupling with an aryl boronic
acid 22 (Scheme 3) that mimics the benzofuranone environ-
ment of 2. Strategic considerations suggested that it would
be desirable to convert the C(16) bromide into a boronic
acid substituent prior to the incorporation of a carboxyl group
at C(10). The main advantage of this plan is that the high
(6) (a) Hollins, R. A.; Colnago, L. A.; Salim, V. M.; Seidl, M. C. J.
Heterocycl. Chem. 1979, 16, 993. (b) Somei, M.; Yamada, F.; Kunimoto,
M.; Kaneko, C. Heterocycles 1984, 22, 797.
(7) Batcho, A. D.; Leimgruber, W. Org. Synth. 1985, 63, 214.
(8) Amat, M.; Hadida, S.; Pshenichnyi, G.; Bosch, J. J. Org. Chem. 1997,
62, 3158.
(9) Alo, B. I.; Kandil, A.; Patil, P. A.; Sharp, M. J.; Siddiqui, M. A.;
Josephy, P. D.; Snieckus, V. J. Org. Chem. 1991, 56, 3763.
(10) For a review of Suzuki coupling reactions, see: Miyaura, N.; Suzuki,
A. Chem. ReV. 1995, 95, 2457.
1034
Org. Lett., Vol. 2, No. 8, 2000

conclude that the two species are the atropisomers corre-
sponding to two possible orientations of the indole ring with
respect to the stereocenter at C(10). The isomeric species
should therefore interconvert thermally by rotation about the
C(16)-C(18) bond. Supporting evidence for the presence
of atropisomers was obtained from NMR experiments
performed in deuterated DMSO in the temperature range
from 70 to 115 °C. The first indications of signal broadening
were detected at 90 °C, and coalescence was seen at 105
and 115 °C for the C(10) methine and methoxy signals,
respectively, corresponding to a free energy of activation of
ca. 19 kcal/mol for the interconversion process.
Scheme 3
Other groups have investigated mechanistically similar
biaryl coupling reactions en route to potential precursors of
diazonamide A,2b-e However, our results represent the first
Suzuki coupling reactions of a 4-trifluoromethylsulfonyl-
oxyindole. It is significant that the hindered triflate 3
participates efficiently in these reactions and that lactone
functionality is compatible with the activation and coupling
sequence. Furthermore, this route allows assembly of the
oxazole prior to the coupling reaction with an advanced,
lactone-containing boronic acid fragment. It also provides
potential flexibility for introduction of a carbon substituent
at diazonamide C(28) or a chloride at diazonamide C(27) at
several stages of the synthesis using lithiooxazole chemis-
try.12 Some progress with the chlorination of a lithiated
oxazole has already been made in preliminary studies.13 The
alternative of electrophilic ring halogenation of an oxazolyl
indole is also available from studies by Wipf et al.2e
In conclusion, a practical method for the preparation of
4-substituted 3-(oxazol-5-yl)indoles has been demonstrated
and shown to be useful in the synthesis of a model compound
corresponding to the biaryl and 3-(oxazol-5-yl)indole seg-
ment of diazonamide A. Future progress toward this end will
be reported in due course.
acidity of the benzofuranone C(10)-H in 20 provides a simple
means for lactone carbonyl protection as the enolate 21
during the conversion of C(16)-Br into an activated organo-
metallic intermediate.
Benzofuranone 20 was prepared using a sequence based
on the synthesis of 23 (Scheme 3; 50% over three steps from
the chloromethyl ketone 1711). Next, protection of the lactone
carbonyl was accomplished by treatment of 20 with 2 equiv
of NaH at 0 °C in THF to generate 21. Subsequent addition
of 2 equiv of n-BuLi at -78 °C and quenching with 3 equiv
of B(OMe)3 gave the boronic acid 22 with minimal damage
to the lactone ring (66% of 22 isolated).
Acknowledgment. This work was supported by NIH
(CA17918) and by the award of an NIH postdoctoral
fellowship to D.A.B.
Supporting Information Available: Experimental pro-
cedures and characterization data for new compounds. This
material is available free of charge via the Internet at
http://pubs.acs.org.
The cross coupling reaction of 3 and 22 was investigated
using the same conditions that had been optimized in the
simpler example (14). A coupling product was isolated in
63% yield, and NMR spectra demonstrated that the crucial
C(16)-C(18) bond had been formed. However, the spectra
also provided clear evidence that two isomeric structures
were present in a 1:1 ratio. Thus, two signals were observed
for the methoxy group at C(17), for the methine hydrogen
at C(10), and also for several of the aromatic C-H bonds.
Because the model compound 15 has a relatively simple
spectrum (singlets for oxazole C-H; methoxy; tert-butoxy
protons), hindered rotation of the oxazole is not a likely
explanation for the doubling of signals observed for 23. We
OL005548P
(12) (a) Vedejs, E.; Monahan, S. D. J. Org. Chem. 1996, 61, 5192. (b)
Vedejs, E.; Luchetta, L. M. J. Org. Chem. 1999, 64, 1011.
(13) According to the method of ref 12b, treatment of the 3-(oxazol-5-
yl)indole 24 with 2.15 equiv n-butyllithium in THF/DMPU at -78 °C
followed by slow addition of N-chlorosuccinimide in dichloromethane and
warming to room temperature gave 25 in 60% yield: Vedejs, E.; Zajac,
M., unpublished results.
(11) Structure 6, preceding paper. See ref 3.
Org. Lett., Vol. 2, No. 8, 2000
1035