Stereoselective synthesis of polyketide fragments using a novel intramolecular Claisen-like condensation/reduction sequence

Article abstract of DOI:10.1016/S0040-4039(01)01840-8

Intramolecular Claisen-type cleavage of the Evans-oxazolidinone with an acetate enolate followed by reduction of the resulting ketone using a borane-amine complex yielded β-hydroxy-δ-lactones as fully functionalized polyketide precursors stereoselectively

Full text of DOI:10.1016/S0040-4039(01)01840-8

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 42 (2001) 8463–8465
Stereoselective synthesis of polyketide fragments using a novel
intramolecular Claisen-like condensation/reduction sequence
Klaus Hinterding,* Suradech Singhanat and Lukas Oberer
Novartis Pharma AG, Transplantation Research, WSJ507.409, CH-4002 Basel, Switzerland
Received 17 September 2001; accepted 1 October 2001
Abstract—Intramolecular Claisen-type cleavage of the Evans-oxazolidinone with an acetate enolate followed by reduction of the
resulting ketone using a borane–amine complex yielded b-hydroxy-d-lactones as fully functionalized polyketide precursors
stereoselectively. Consequently, this reaction sequence constitutes a highly practical alternative to an acetate–aldol reaction.
© 2001 Elsevier Science Ltd. All rights reserved.
Chiral polyketide units are important fragments of
many natural products, especially macrolide antibi-
otics,1 ionophores2 and the immunosuppressive natural
products FK506,3 Rapamycin4 and Sanglifehrin A.5 In
order to investigate derivatives of immunosuppressants
with novel modes of action, we needed an efficient
synthesis of unknown polyketide fragments 1 and 2
(Fig. 1).
ting the stereochemistry at C3 requires a non-trivial
selective acetate–aldol reaction (or equivalent thereof),
for which novel synthetic methodologies are still
urgently needed.8 Indeed, initial studies with nucle-
ophilic additions of acetate enolates and allyl-Ti
reagents to aldehydes 3 and 4 were met by limited
success. Therefore, we reasoned that an intramolecular
cleavage of the Evans-auxiliary by an acetate enolate to
form the d-lactones 5 and 6,9 followed by a stereoselec-
tive reduction of the ketone would constitute an attrac-
tive alternative to the acetate aldol reaction. This
synthetic sequence obviates the need to liberate 3 and 4
from the initial aldol adducts and to perform protecting
group manipulations. Herein we report the successful
implementation of this strategy.
Using the Evans-oxazolidinone, the stereocenters at C4
and C5 can be established efficiently in syn-6 and
anti-selective7 propionate aldol reactions. However, set-
As shown in Scheme 1, anti aldol 710 was transformed
into the corresponding acetate 8 by treatment with
acetic anhydride in the presence of DMAP and NEt3.
No epimerization occurred during this step, as detected
by NMR. Exposure of compound 8 to a solution of
LiHMDS (Li-bistrimethylsilylamide) in THF at −78°C
for 3 h led to the clean formation of b-keto-d-lactone
5,11 with the free oxazolidinone as the only detectable
by-product.12 Acidic lactone 5 was isolated conve-
niently by a simple extraction procedure, which also
allowed for efficient recovery of the chiral auxiliary.
Chemoselective reduction of the keto-function in 5
turned out to be troublesome, since conventional reduc-
ing agents either afforded enolization only or complex
reaction mixtures.13 Clean conversion of ketone 5 into
secondary alcohol 9,14 however, was achieved using
t-BuNH2-BH3 in combination with citric acid.15 The
ratio of C3 epimers was determined by NMR to be
equal to 13/1. NOE experiments revealed that relative
Figure 1.
Keywords: acetate–aldol reaction; polyketides; Evans-auxiliary; b-
keto-d-lactones.
* Corresponding author. Tel.: +41613243724; fax: +41613246735;
e-mail: klaus.hinterding@pharma.novartis.com
0040-4039/01/$ - see front matter © 2001 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(01)01840-8

8464
K. Hinterding et al. / Tetrahedron Letters 42 (2001) 8463–8465
Scheme 1. Reagents and conditions: (a) Ac2O, NEt3, DMAP, CH2Cl2, rt; (b) LiHMDS, THF, −78°C; (c) t-BuNH2·BH3, citric acid,
MeOH, H2O, −5 to 0°C; (d) LiOH, H2O, THF.
stereochemistry and conformation of the major
stereoisomer are as shown in Fig. 2: lactone 9 adopts a
boat-conformation with all substituents in equatorial
positions.
present in chair- and boat-like conformations of keto-
lactone 5. Treatment of lactone 9 with one equivalent
of LiOH in THF/H2O completed the synthesis of
stereotriad 1 as its Li-salt. The same reaction sequence
proved effective to transform syn-aldol 1017 into b-
hydroxy-d-lactone 12 and subsequently into carboxylate
2 (Scheme 2). Again the reduction of keto-lactone 6
yielded one major isomer of 12.18 Its relative stereo-
chemistry was determined by NOE experiments to be as
shown in Scheme 2. The face selectivity can be rational-
ized by the fact that either the ethyl- or methyl-sub-
stituent in lactone 6 is forced into an axial position and
prevents attack of the reducing agent.
Although the exact conformation of ketone 5 is
unknown,16 the relatively high facial selectivity is pre-
sumably the consequence of steric hindrance of two
axial protons next to the reacting carbonyl center
In summary, we described a novel and highly practical
alternative to a selective acetate–aldol reaction for the
synthesis of polyketide fragments. The conversion of
syn- and anti-aldols 7 and 10 into stereotriads 1 and 2
can be achieved in only four steps and proceeds in good
overall yield (44–60%).
Acknowledgements
We wish to thank Gebhard Thoma and Rolf Brein-
bauer for helpful discussions.
Figure 2. Rationalization of stereoselective reduction and
observed NOEs in boat conformation of 9.
Scheme 2. Reagents and conditions: (a) Ac2O, NEt3, DMAP, CH2Cl2, rt; (b) LiHMDS, THF, −78°C; (c) t-BuNH2·BH3, citric acid,
MeOH, H2O, −5 to 0°C; (d) LiOH, H2O, THF.

K. Hinterding et al. / Tetrahedron Letters 42 (2001) 8463–8465
8465
References
Leijonmarck, H. J. Chem. Soc., Chem. Commun. 1983,
1097–1098.
1. O’Hagan, D. Nat. Prod. Rep. 1995, 12, 1–32.
2. Dutton, C. J.; Banks, B. J.; Cooper, C. B. Nat. Prod.
Rep. 1995, 12, 165–181.
3. Tanaka, H.; Kuroda, A.; Marusawa, H.; Hatanaka, H.;
Kino, T.; Goto, T.; Hashimoto, M.; Taga, T. J. Am.
Chem. Soc. 1987, 109, 5031–5033.
4. (a) Vezina, C.; Kudelski, A.; Sehgal, S. N. J. Antibiot.
1975, 28, 721–726; (b) Sehgal, S. N.; Baker, H.; Vezina,
C. J. Antibiot. 1975, 28, 727–732.
12. This result is remarkable in the light of Branda¨nges
findings (see Ref. 9). In a related experiment using a
Norephedrin-derived oxazolidinone he observed predomi-
nant attack at the endo-carbonyl function, yielding the
11-membered ring structure. Subtle effects of substituents
on the oxazolidinone therefore seem to govern the course
of the reaction.
5. (a) Sanglier, J. J.; Quesniaux, V.; Fehr, T.; Hofmann, H.;
Mahnke, M.; Memmert, K.; Schuler, W.; Zenke, G.;
Gschwind, L.; Maurer, C.; Schilling, W. J. Antibiot. 1999,
52, 466–473; (b) Fehr, T.; Kallen, J.; Oberer, L.; Sanglier,
J. J.; Schilling, W. J. Antibiot. 1999, 52, 474–479.
6. Evans, D. A. Aldrichim. Acta 1982, 15, 23–32.
7. Raimundo, B. C.; Heathcock, C. H. Synlett 1995, 1213–
1214.
8. For detailed information on this topic, see: Braun, M. In
Stereoselective Synthesis (Houben-Weyl); Helmchen, G.;
Hoffmann, R. W.; Mulzer, J.; Schaumann, E., Eds.;
Thieme: Stuttgart, 1996; Vol. E21b, pp. 1603–1666.
9. Branda¨nge, S.; Leijonmarck, H. Tetrahedron Lett. 1992,
33, 3025–3028.
13. K-Selectride and DIBAH led to enolization, LiBH4 and
NaBH4 in solvents such as MeOH, THF to complex
reaction mixtures detected by TLC, while the use of
Me3N-BH3 gave no conversion.
14. Experimental procedure: To a stirred solution of ketone 5
(35 mg, 0.22 mmol) in MeOH (2 ml) was added at −5°C
a solution of t-BuNH2-BH3 (62 mg, 0.67 mmol) in
MeOH (2 ml), immediately followed by a 1 M solution of
citric acid in water (1.6 ml). After 2.5 h at −5°C, CH2Cl2
(15 ml) and water (5 ml) were added. Layers were sepa-
rated and the aqueous phase was extracted with CH2Cl2
(3×10 ml). After drying with MgSO4 and concentration
of the organic layer, chromatography (ethylacetate/hex-
ane=2/1) yielded 26 mg (0.16 mmol, 73%) of a colorless
10. Prepared according to Ref. 7.
11. Experimental procedure: A solution of acetate 4 (1.2 g,
3.2 mmol) in THF (25 ml) was added at −78°C to a
stirred solution of LiHMDS (1 M in THF, 9.6 ml, 9.6
mmol) in THF (20 ml). After stirring for 3 h, the reaction
mixture was quenched at −78°C by the addition of satd
NH4Cl/MeOH/H2O (100 ml, 1/1/1, v/v/v). Ethylacetate
(100 ml) and water (30 ml) were added and the layers
separated. The organic phase contained the chiral auxil-
iary in quantitative yield and 90% purity. The aqueous,
basic layer (pH 9–10) was titrated with 0.3N HCl to pH
2–3 and then extracted with CH2Cl2 (3×100 ml). Drying
with MgSO4 and concentration of the organic layer
yielded the product as an off-white solid in quantitative
yield and 90% purity. Chromatography (ethylacetate/hex-
ane=1/2) gave pure 5 as a white solid (485 mg, 3.1 mmol,
95%): Rf=0.35 in ethylacetate/hexane=1/2 mp: 78–79°C;
1H NMR (400 MHz, CDCl3): l=1.12 ppm (t, J=7 Hz,
1
oil. Rf=0.21 in ethylacetate/hexane=1/1; H NMR (500
MHz, DMSO): l=0.90 ppm (d, J=7.4 Hz, 3H,
CHCH
6
3), 0.94 (t, J=6.8 Hz, 3H, CH2CH3
6 ), 1.47 (m, 2H,
CH2CH3), 1.70 (m, 1H, CH
6
6
CH3), 2.25 (dd, J=6.4 Hz,
J=16.5 Hz, 1H, CH2CO), 2.78 (dd, J=5.6 Hz, J=16.5
Hz, 1H, CH2CO), 3.60 (m, 1H, CHOCO), 3.85 (ddd,
J=2.9 Hz, J=7.6 Hz, J=10.4 Hz, 1H, CH6 OH), 5.12 (d,
J=4.7 Hz, 1H, OH); MS (ES+): 159 (MH+), 141 (M+−
H2O), 129 (M+−C2H5), 116, 87, 58; HRMS calcd for
C8H14O3+OH 175.0970, found 175.0971; IR: 3430 (m,
OH), 2971, 2937, 2883 (m, CH), 1729 (s, lacton CO),
1462 (m), 1379, 1360 (m), 1250 (s, lacton CO), 1190,
1098, 1042, 999 (m); [h]2D5=+41.9, c=8 mg/ml in Et2O.
15. Bennett, F.; Fenton, G.; Knight, D. W. Tetrahedron
1994, 50, 5147–5158.
3H, CH2CH3
(ddq, J=8 Hz, 1H, CH
J=8 Hz, 1H, CH2CH3), 2.43 (dq, J=10 Hz, J=7 Hz,
1H, CHCH3), 3.45 (d, J=19 Hz, 1H, CH2CO), 3.56 (d,
J=19 Hz, 1H, CH2CO), 4. 28 (ddd, J=10 Hz, J=7 Hz,
6
), 1.19 (d, J=7 Hz, 3H, CHCH3
6 ), 1.68–1.77
16. The degree of enolization of keto-lactone 5 adds to the
difficulty of predicting its reacting conformation in solu-
tion. A CDCl3 solution showed 5 as its keto-tautomer
(>95%), while in DMSO the enol-form prevailed (>90%).
17. Gage, J. R.; Evans, D. A. Org. Synth. 1989, 68, 77–91.
18. The ratio of C3 epimers was detected by NMR to be
equal to 4/1. Chromatographic separation of epimers
gave pure 10 in 60% yield.
6
2CH3), 1.92–2.00 (ddq, J=3 Hz,
6
6
6
6
J=3 Hz, 1H, CHO); MS (EI): 156 (M+), 127 (M−C2H5+),
98, 85, 56, 42; HRMS calcd for C8H12O3 155.0708, found
155.0710; [h]2D5=+144, c=5.5 mg/ml in Et2O. NMR data
are in accordance with that of racemic 5: Branda¨nge, S.;