Synthesis of chiral and highly functionalized cyclopentanes

Article abstract of DOI:10.1016/S0040-4039(00)02320-0

The unexpected reduction of the acetal moiety at the glycosidic C1-O-exo bond in the bis-heteroannulated-pyranosides 3 and 9 is a simple and new method for the synthesis of chiral and highly functionalized cyclopentanes.

Full text of DOI:10.1016/S0040-4039(00)02320-0

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 42 (2001) 1515–1517
Synthesis of chiral and highly functionalized cyclopentanes†
Jose´ Marco-Contelles* and Juliana Ruiz-Caro
Laboratorio de Radicales Libres (CSIC), C/Juan de la Cierva, 3, 28006 Madrid, Spain
Received 30 October 2000; accepted 16 December 2000
Abstract—The unexpected reduction of the acetal moiety at the glycosidic C1ꢀO-exo bond in the bis-heteroannulated-pyranosides
3 and 9 is a simple and new method for the synthesis of chiral and highly functionalized cyclopentanes. © 2001 Elsevier Science
Ltd. All rights reserved.
For years we have been very interested in the synthesis
of chiral, highly functionalized cyclopentanes via free
radical mediated carbocyclization of conveniently func-
tionalized precursors derived from sugars.1 Simulta-
Boom and co-workers reported on the synthesis of
structural fragments of type B (Fig. 1).6 This prompted
us to describe here the unexpected results obtained in
the reduction of the acetal moiety of the glycosidic
C1ꢀO-exo bond in the complex adducts (3 and 9).
Compound 3 was obtained by PK reaction of the
neously, we have also prepared
a
series of
cyclopenta[c]pyrans by Pauson–Khand (PK) reaction2
on pyranoside templates derived from carbohydrates.3
This structural motif (A, Fig. 1) is widespread in nature
in biologically interesting molecules.4 As a consequence,
a number of strategies have been devised for the synthe-
sis of these natural products.5 In a recent paper van
2-(prop-2-ynyl)-2,3-dideoxy-a-D-erythro-hex-2-enopyra-
noside (1); product 9 was prepared by manipulation of
the PK reaction product 4, synthesized from the corre-
sponding 2-(prop-2-ynyl)-2,3-dideoxy-a-
D-erythro-hex-
2-enopyranoside (2).
Figure 1.
Scheme 1.
* Corresponding author. Tel.: 34 91 562 29 00; fax: 34 91 564 48 53; e-mail: iqoc21@iqog.csic.es
†
Dedicated to Professor Joachim Thiem on the occasion of his 60th birthday.
0040-4039/01/$ - see front matter © 2001 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(00)02320-0

1516
J. Marco-Contelles, J. Ruiz-Caro / Tetrahedron Letters 42 (2001) 1515–1517
This protocol resulted in a new strategy for the synthe-
sis of chiral, polyfunctionalized cyclopentanes of type C
(Fig. 1), difficult to obtain by other procedures, and
with wide potential synthetic interest.
of product 6 from 3 is probably due to the presence of
the sterically hindered isopropylidene acetal that pre-
vents the attack at C12. However, it is well known that
the regioselective cleavage of benzylidene acetals gives
benzyl ethers using AlCl3/LiAH9 or NaCNBH3/HCl.10
The use of DIBALH for such a purpose has also been
described.11
In the course of our ongoing project for the synthesis of
iridoids via PK reaction on carbohydrates,3e we consid-
ered the reduction of ketones 3 and 4. These com-
pounds were prepared by PK reaction of
Continuing with our project, the benzylidene protected
compound 4, under the same experimental conditions
(DIBALH, 2.0 equiv., 1.0 M in toluene, at −78°C; in
toluene as solvent), afforded adduct 8 in 89% yield
(Scheme 1). When the reduction was performed in
THF, under the same conditions, compound 8 was
isolated in 73% yield. Product 8 was transformed into
intermediate 98 by standard conditions (t-
butyldimethylsilyl chloride, imidazole, methylene chlo-
ride, rt) in 87% yield (Scheme 2).
2-(prop-2-ynyl)-2,3-dideoxy-a-D-erythro-hex-2-enopyra-
nosides 1 and 2, respectively (Scheme 1). It is interest-
ing to point out that these compounds, having a
1,3-dioxane ring involving the oxygens at positions C6
and C4 in the pyranoside template, have given the best
yields so far obtained by us for the PK reactions in
sugar templates.7
We have found that the reduction of the bis-heteroan-
nulated-pyranoside 3 with DIBALH (2.0 equiv., 1.0 M
in toluene), at −78°C, in toluene, afforded the allylic
alcohols 58 and 6,8 in 69 and 9% yields, respectively
(Scheme 2). In both cases the ketone reduction pro-
ceeded stereoselectively from the less hindered b-face to
give the expected compounds with R configuration at
C6,3 as determined by full spectroscopic analysis. The
formation of product 6,8 albeit in low yield (9%),
during the reduction of product 3, was unexpected.
Acetylation of product 6 gave a diacetate (7),8 whose
structure was confirmed by analytical and spectroscopic
data. Particularly significant in this product (7) was the
absence of the anomeric proton (H-1), which was sub-
In order to test the benzylidene hydrogenolysis versus
the C1ꢀO glycosidic cleavage, we submitted compound
9 to reduction (Scheme 3). From compound 9, with
toluene as solvent and at −78°C,12 alcohol 10 (26%
(41%)) was the only isolated product; when the reaction
was performed at −40°C, the regiochemistry was the
same but the yield was better (57% (95%)). Repeating
the process using methylene chloride, at −78°C,12 alco-
hol 10 (21% (43%)) was obtained; when the reaction
was performed at −40°C, a mixture of products 10
(79%) and 11 (8%) resulted. The analytical and spectro-
scopic data of compound 10 clearly showed this was
again the result of the exclusive and regioselective
cleavage of the glycosidic C1ꢀO-exo bond. In fact, as in
compounds 6/7 (Scheme 2), we readily noticed the
absence of the anomeric proton, which was substituted
by two new protons at 4.14 ppm (dd, 1H, J1a,1b=11.0
Hz, J1b,9=6.1 Hz, H-1b) and at 3.56 ppm (t, 1H,
stituted by two protons at 4.17 ppm (ddd, 1H, J1a,1b
=
11.5 Hz, J1b,9=6.6 Hz, J=0.5 Hz, H-1b) and at 3.38
ppm (t, 1H, J1a,1b=J1a,9=11.5 Hz, H-1a), that led us to
conclude that the glycosidic C1ꢀO-exo bond was par-
tially broken in the reduction of compound 3 to yield
minor, but significant quantities of product 5. Based on
the chemical and spectroscopic evidence, the formation
of product having a secondary alcohol, after cleavage
of the C1ꢀO-endo bond, was eliminated. The formation
J
1a,1b=J1a,9=11.0 Hz, H-1a). Finally, peracetylation of
alcohol 11 afforded product 128 (7% overall yield from
Scheme 2.

J. Marco-Contelles, J. Ruiz-Caro / Tetrahedron Letters 42 (2001) 1515–1517
1517
Scheme 3.
9) (Scheme 3). Substrate 1113 was the result of the
reduction of the benzylidene group plus the reduction
of the C1ꢀO-exo bond.
5. Lorenzo, E.; Alonso, F.; Yus, M. Tetrahedron Lett. 2000,
41, 1661 and references cited therein.
6. Leeuwenbugh, M. A.; Appeldoorn, C. C. M.; van Hooft,
P. A. V.; Overkleeft, H. S.; van der Marel, G. A.; van
Boom, J. H. Eur. J. Org. Chem. 2000, 873.
Overall, the reduction of the acetal at the glycosidic
C1ꢀO-exo bond, in molecules 3 and 9, is worthy of note
and unexpected. This is probably a consequence of the
high tension in these bis-heteroannulated-pyranosides,
coupled with the geometry of these products which
present b-faces at C-1 free of steric constraints for any
attacking reagent, as we can see by simple inspection of
molecular models.
7. A large series of differently substituted 2-(prop-2-ynyl)-
2,3-dideoxy-a-D-erythro-hex-2-enopyranosides at C4-O
and C6-O positions have been synthesized and submitted
to PK reaction, showing the critical and important effects
of these groups on the yields and in the structure of the
resulting adducts (Marco-Contelles, J.; Ruiz-Caro, J., to
be published).
8. All new compounds showed excellent analytical and spec-
1
In summary, a series of unexpected results have
afforded a new and simple protocol for the synthesis of
chiral and highly polyfunctionalized cyclopentanes for
further synthetic elaboration. Work is now in progress
in our laboratory in order to explore the synthetic
scope of this hydride-mediated regioselective glycosidic
reduction. The results will be reported in due course.
troscopic data. Selected spectroscopic data. 10: H NMR
(300 MHz, CDCl3) l 7.54–7.34 (m, 5H, C6H5), 5.86 (br s,
1H, H-7), 5.58 (s, 1H, H-12), 4.82 (dd, 1H, J6,5=6.0 Hz,
J
6,7=2.0 Hz, H-6), 4.33 (dd, 1H, J11%,11=10.0 Hz, J11%,3
=
=
5.0 Hz, H-11%), 4.24 (td, 1H, J3,4=J3,11=10.0 Hz, J3,11%
5.0 Hz, H-3), 4.20 (br s, 2H, 2H-10), 4.14 (dd, 1H,
J
1%,1=11.0 Hz, J1%,9=6.0 Hz, H-1%), 3.96 (dd, 1H, J4,3
=
=
10.0 Hz, J4,5=6.0 Hz, H-4), 3.59 (t, 1H, J11,11%=J11,3
10.0 Hz, H-11), 3.56 (t, 1H, J1,1%=J1,9=11.0 Hz, H-1),
2.91 (dt, 1H, J9,1=11.0 Hz, J9,1%=J9,5=6.0 Hz, H-9), 2.56
(q, 1H, J5,6=J5,4=J5,9=6.0 Hz, H-5), 1.59 (br s, 1H,
OH), 0.89 [s, 9H, -C(CH3)3], 0.04 (s, 3H, -Si-CH3), 0.03
(s, 3H, -Si-CH3); 13C NMR (75 MHz, CDCl3) l 149.1
(C-8), 129.7 (C-7), 137.8–126.4 (C6H5), 102.6 (C-12), 80.1
(C-4), 75.1 (C-6), 71.7 (C-1), 70.3 (C-11), 68.8 (C-3), 61.1
(C-10), 45.3 (C-9)*, 43.9 (C-5)*, 25.9 [-C(CH3)3], 17.9
[-C(CH3)3], −4.5 (-Si-CH3), −4.8 (-Si-CH3) (* these values
can be interchanged).
Acknowledgements
J.R.C. thanks Consejer´ıa de Educacio´n y Cultura
(CAM) for a pre-doctoral fellowship. J.M.C. thanks
CICYT (Petri, BQU2000-1495), CAM and EU (COST
Action D-12) for continued and generous support.
References
9. Bhattacharjee, S. S.; Gorin, P. A. J. Can. J. Chem. 1969,
47, 1195.
1. (a) Marco-Contelles, J.; Mart´ınez-Grau, A. Chem. Soc.
Rev. 1998, 27, 155; (b) Marco-Contelles, J.; Alhambra,
C.; Mart´ınez-Grau, A. Synlett 1998, 693 (Corrigendum:
Synlett 1999, 376).
2. (a) Schore, N. E. Org. React. 1991, 40, 1; (b) Marco-Con-
telles, J.; Ingate, S. Org. Prep. Proc. Int. 1998, 30, 121; (c)
Brummond, K. M.; Kent, J. L. Tetrahedron 2000, 56,
3263.
3. (a) Marco-Contelles, J. Tetrahedron Lett. 1994, 35, 5059;
(b) Marco-Contelles, J. J. Org. Chem. 1996, 61, 7666; (c)
Marco-Contelles, J.; Ruiz, J. Tetrahedron Lett. 1998, 39,
6393; (d) Marco-Contelles, J.; Ruiz, J. J. Chem. Res. (S)
1999, 160; (e) Marco-Contelles, J.; Ruiz-Caro, J. J. Org.
Chem. 1999, 64, 8302.
10. Garegg, P. Pure Appl. Chem. 1984, 56, 845.
11. Mikami, T.; Asano, H.; Mitsunobu, O. Chem. Lett. 1987,
2033.
12. In a typical experiment, product (9) was dissolved in
dry toluene (or CH2Cl2) (0.15 M) and cooled at −78
(or −40°C) under argon and stirring. DIBALH (3 equiv.,
1.0 M in toluene) was added. After 4 h, DIBALH (2
equiv., 1.0 M in toluene) was added again to the mixture.
After 4 h the reaction was quenched with methanol and
warmed at rt. The salts were filtered over Celite-545, and
the filtrate was evaporated to give a residue which was
submitted to chromatography.
13. Compound 11 presented the benzyl group at O-C4, in
agreement with similar trends observed for the reduction
of the benzylidene groups with AlCl3/LiAlH4 (see Ref. 9).
4. El-Naggar, L. J.; Beil, J. L. J. Nat. Prod. 1980, 42, 649.
.