Studies on the synthesis of the mycalamides: Stereocontrolled synthesis of a model N-glycosylpederamide via a stereoselective aldol reaction

Full text of DOI:10.1021/jo980023k

2064
J . Org. Chem. 1998, 63, 2064-2065
Sch em e 1a
Stu d ies on th e Syn th esis of th e Myca la m id es:
Ster eocon tr olled Syn th esis of a Mod el
N-Glycosylp ed er a m id e via a Ster eoselective
Ald ol Rea ction
William R. Roush,*,1 Lance A. Pfeifer, and
Thomas G. Marron
Department of Chemistry, Indiana University,
Bloomington, Indiana 47405, and Department of Chemistry,
University of Michigan, Ann Arbor, Michigan 48109
Received J anuary 7, 1998
Our original plan for completion of total syntheses2,3 of
mycalamides A (1) and B (2)4,5 called for N-acylation6 of
carbamate 37 with a suitable active ester of pederic acid (see
4a -d ).8 Unfortunately, repeated attempts to accomplish
this key transformation by treating either the lithium or
potassium anions generated from 3 (via deprotonation with
n-BuLi, LHMDS, LHMDS-HMPA, or KHMDS in THF at
-78 °C) with the acid chloride 4b or acid fluoride 4c6 under
a range of conditions (e.g., in the presence of added DMPU
or HMPA) and reaction temperatures (-78 to 0 °C) gave
none of the desired coupled product.9 Surprisingly, both 4b
and 4c could be recovered chromatographically from the
reaction mixtures. The lack of success with this reaction
can be attributed to severe nonbonded interactions that
develop in tetrahedral intermediate 6, which experiences two
destabilizing gauche pentane interactions10 no matter which
rotamer about the C(6,7) or C(7,8) bonds is considered. The
low reactivity of pederoyl chloride derivatives has been noted
previously.11 Attempts were also made to deprotect 3 (via
treatment with TBAF in DMF) and to perform the acylation
of amine 5 with the mixed anhydride 4d according to Kishi’s
protocol.2a This, however, provided enal 7 as the major
product, again with none of the desired amide being iso-
lated.12
a
Key: (a) butyl vinyl ether, Hg(OAc)2, 70%; (b) LiOH, H2O2, THF-
H2O, 84%; (c) OsO4, NMO, t-BuOH-THF-H2O, 69%.
to probe this strategy by using glucosylamine derivative 9
as a model system. We report herein the results of these
investigations, culminating in a highly stereoselective syn-
thesis of the N-glucosylpederamide derivative 22.
Carbamate 9 was prepared from the known tetraacetyl
glycosyl azide 8,13 while carboxylic acid 11 was prepared
from the previously described pederic acid precursor 10
(Scheme 1).8 Unfortunately, acid 11 also proved to be too
hindered to undergo efficient coupling with 9. Best results
were obtained when 11 was converted to the mixed phos-
phinic anhydride by sequential treatment with n-BuLi in
THF (-78 °C) followed by addition of diphenylphosphinic
chloride. Addition of a solution of this active ester (1.3 equiv)
to a -78 °C solution of the lithium anion generated from 9
(LiHMDS, THF, -78 °C, in the presence of 4 Å molecular
sieves) with warming to ambient temperature provided 12
in 21% yield along with 63% of recovered 9. Comparable
yields of 12 were obtained with the acid chloride generated
from 11.6 Less reproducible results were obtained when the
lithium anion of 9 was treated with the mixed anhydride
generated from 11 and trichlorobenzoyl chloride (10-30%
of 12; 33-42% of recovered 9). Attempts to use â-lactone
14 as the acylating agent were unsuccessful.
(2) Total syntheses of mycalamides A and B: (a) Hong, C. Y.; Kishi, Y.
J . Org. Chem. 1990, 55, 4242. (b) Nakata, T.; Fukui, H.; Nakagawa, T.;
Matsukura, H. Heterocycles 1996, 42, 159. (c) Synthesis of 18-O-methyl
mycalamide B: Kocienski, P.; Raubo, P.; Davis, J . K.; Boyle, F. T.; Davies,
D. E.; Richter, A. J . Chem. Soc., Perkin Trans. 1 1996, 1797.
(3) Other synthetic studies on the mycalamides: (a) Hoffmann, R. W.;
Schlapbach, A. Tetrahedron Lett. 1993, 34, 7903. (b) Hoffmann, R. W.;
Breitfelder, S.; Schlapbach, A. Helv. Chim. Acta 1996, 79, 346.
(4) Perry, N. B.; Blunt, J . W.; Munro, M. H. G.; Thompson, A. M. J . Org.
Chem. 1990, 55, 223.
(5) Perry, N. B.; Blunt, J . W.; Munro, M. H. G.; Pannell, L. K. J . Am.
Chem. Soc. 1988, 110, 4850.
(6) Roush, W. R.; Pfeifer, L. A. J . Org. Chem. 1998, 63, 2062.
(7) Marron, T. G.; Roush, W. R. Tetrahedron Lett. 1995, 36, 1581.
(8) Roush, W. R.; Marron, T. G.; Pfeifer, L. A. J . Org. Chem. 1997, 62,
474.
These results dictated that we examine a revised approach
involving use of less advanced, less sterically congested
pederic acid precursors in the carbamate acylation reaction.
Because our supply of 3 was virtually exhausted, we elected
(1) Correspondence to this author should be sent to the University of
Michigan address.
(9) Marron, T. G. Ph.D. Thesis, Indiana University, 1995.
(10) Hoffmann, R. W. Angew. Chem., Int. Ed. Engl. 1992, 31, 1124.
S0022-3263(98)00023-1 CCC: $15.00 © 1998 American Chemical Society
Published on Web 03/17/1998

Communications
J . Org. Chem., Vol. 63, No. 7, 1998 2065
Sch em e 2a
prepared by a standard sequence of operations from 17.8
Interestingly, this reaction provided a single aldol diaste-
reomer 19 in 61% yield; in addition, 25% of aldehyde 18 and
70% of 16 were recovered. The stereochemistry of 19 was
assigned by analogy to 13, which was the only diastereomer
produced in the chlorotitanium aldol reaction of 23 and
aldehyde 24.15 The stereochemistry of 13 is known unam-
biguously by virtue of its synthesis from 12 (vide supra). The
aldol reaction of 24 and the lithium enolate of 23 was also
highly stereoselective, although in this case it was not
possible to suppress migration of the Teoc group from
nitrogen to the resulting aldolate oxygen. No reaction
occurred under standard boron aldol conditions (Bu2BOTf,
Et3N, CH2Cl2, -78 to 0 °C). The excellent diastereoselec-
tivity of these reactions appears to be due the tendency of
â-alkoxy aldehydes to favor the generation of 1,3-anti
products16 and not due to a high diastereofacial bias on the
part of the metal enolate,17 since aldol reactions of 16 with
achiral aldehydes provided ca. 2:1 mixtures of products.
Swern oxidation18 of 19 gave a sensitive â-keto imide that
was immediately treated with camphorsulfonic acid (CSA)
in MeOH. This initiated a sequence of three functional
group transformationssdeprotection of the two TBS ethers
and cyclization of the δ-hydroxy ketone to the methyl
hemiketal unit of 20, which was isolated in 67% yield for
the two steps. Significantly, no epimerization of the C(7)
stereocenter occurred under these conditions. Oxidation of
20 under Swern conditions followed by introduction of the
exo-methylene unit by using the Takai-Nozaki protocol
(CH2I2, TiCl4, Zn, THF)19 provided the fully protected
pederamide derivative 21 in 76% yield. Finally, removal of
the Teoc protecting group by treatment of 21 with n-Bu4NF
in DMF at 0 °C and the benzyl ether by dissolving metal
reduction (Na, NH3) completed the synthesis of the model
N-glycosylpederamide 22.
a
Key: (a) Na, NH3, -78 °C, 93%; (b) TBS-OTf, 2,6-lutidine, 92%;
(c) O3, CH2Cl2-MeOH, -78 °C, then Ph3P, 86%; (d) DMSO, (COCl)2,
Et3N, -78 °C; (e) CSA, MeOH, 67% for two steps; (f) DMSO, (COCl)2,
Et3N, -78 °C, 88%; (g) Zn, CH2I2, TiCl4, THF, 86%; (h) TBAF, DMF,
0 °C, 89%; (i) Na, NH3, -78 °C, 74%.
Ultimately, a workable synthesis of the N-glucosyl ped-
eramide derivative 22 was developed, involving the aldol
reaction of 16 and 18 (Scheme 2). Acylation of 9 with
benzyloxyacetyl chloride (15) provided imide 16 in 86% yield.
This intermediate (3 equiv) underwent a TiCl4-mediated
aldol condensation14 with aldehyde 18, which in turn was
(11) Matsuda, F.; Tomiyoshi, N.; Yanagiya, M.; Matsumoto, T. Tetrahe-
dron 1988, 44, 7063.
In summary, a highly stereoselective method for synthesis
of N-R-alkoxy pederamide derivatives has been devised. This
is only the second3b completely stereocontrolled synthesis
of a pederamide derivative yet reported.20 Further efforts
to streamline this protocol and to utilize it to complete total
syntheses of the mycalamides are in progress and will be
reported in due course.
(12) Treatment of 3 with TBAF in DMF at 0 °C followed by addition of
benzoyl chloride and DMAP provided the N-benzoylmycalamine derivative
in 34% yield as a single diastereomer. In contrast, acylation of the lithium
anion of 3 with benzoyl chloride followed by TBAF removal of the Teoc unit
provided the same N-benzoylmycalamine derivative in 87% yield (ref 7).
(13) Paulsen, H.; Gyo¨rgydea´k, Z.; Friedmann, M. Chem. Ber. 1974, 107,
1568.
(14) Evans, D. A.; Rieger, D. L.; Bilodeau, M. T.; Urpi, F. J . Am. Chem.
Soc. 1991, 113, 1047.
(15) Additional products were obtained resulting from loss of the DMPM
ether protecting group.
(16) Evans, D. A.; Dart, M. J .; Duffy, J . L.; Yang, M. G. J . Am. Chem.
Soc. 1995, 117, 2.
(17) Masamune, S.; Choy, W.; Petersen, J . S.; Sita, L. R. Angew. Chem.,
Int. Ed. Engl. 1985, 24, 1.
(18) Tidwell, T. T. Synthesis 1990, 857.
(19) Hibino, J .; Okazoe, T.; Takai, K.; Nozaki, H. Tetrahedron Lett. 1985,
26, 5579.
Ack n ow led gm en t. This research was supported by
grants from the National Institutes of Health (GM 38907
and RR 10537).
Su p p or tin g In for m a tion Ava ila ble: Experimental proce-
dures and full characterization data for all new compounds; 1H
NMR spectra of 18, 21, and 22 (16 pages).
(20) Leading references to all previous syntheses of pederic acid and
pederamide derivatives are provided in ref 9.
J O980023K