Iminosugars from α,β-epoxyamides. Part 2: Synthetic approach to hydroxylated pyrrolidine and azepane derivatives

Article abstract of DOI:10.1016/j.tetlet.2004.01.144

This paper describes a new synthetic route towards hydroxylated pyrrolidines and azepane derivatives starting from chiral epoxyamides. The alternate transformations on a ribose derivative α,β-epoxyamide permit the construction of five or seven member rings leading to different precursors of homoiminosugars.

Full text of DOI:10.1016/j.tetlet.2004.01.144

Tetrahedron
Letters
Tetrahedron Letters 45 (2004) 2611–2613
Iminosugars from a,b-epoxyamides. Part 2: Synthetic approach
to hydroxylated pyrrolidine and azepane derivatives
z
*
ꢀ
C. Assiego, M. S. Pino-Gonzalez and F. J. Lopez-Herrera
ꢀ
Departamento de Bioquꢀımica, Biologꢀıa Molecular y Quꢀımica Orgꢀanica, Facultad de Ciencias,
Universidad de Mꢀalaga, 29071 Mꢀalaga, Spain
Received 16 December 2003; revised 26 January 2004; accepted 28 January 2004
Abstract—This paper describes a new synthetic route towards hydroxylated pyrrolidines and azepane derivatives starting from
chiral epoxyamides. The alternate transformations on a ribose derivative a,b-epoxyamide permit the construction of five or seven
member rings leading to different precursors of homoiminosugars.
Ó 2004 Elsevier Ltd. All rights reserved.
Homoiminosugars, or iminosugars with an additional
R
6
N
hydroxymethyl group, have received considerable
attention in the last few years because of their remark-
able biological activities.1 Among their useful properties
is their greater selectivity in biological assays in com-
parison to their iminosugar homologues.2 As a result of
their potential therapeutic applications, more efficient
synthesis and/or more selective compounds have been
investigated and many synthetic1–3 and natural4–7 homo-
iminosugars have been identified.
2
CONEt2
OR´´
2
R´
8
O
a
[N]
6
O
b
TrO
OH
O
H
CONEt2
2
[N]
a
O
O
1
R´
N
6
3
We have reported in a recent communication8 a syn-
thetic approach to homoiminosugar derivatives from
a,b-epoxyamides.9 Two strategies (a and b) were plan-
ned starting from the epoxyamide 19a (Scheme 1), and
the formation of piperidine derivatives was described.8
b
TrO
CONEt2
OR
O
O
3
In this paper, we report the development of the second
retrosynthetic alternative (b) with the formation of
pyrrolidine derivatives of the type depicted by 3. These
compounds are aza-C-glycosides10 and can be consid-
ered as precursors or derivatives of homoiminosugars.
Thus, the deprotection and functional group intercon-
version of 3 could afford (inter alia) 2,5-dideoxy-2,5-
H
N
HO
CH2OH
OH
OH
HO
4
imino-D-glycero-D-galacto-heptitol 4. Polyhydroxylated
Scheme 1.
alkaloids with analogous structures have recently been
isolated from plants.11
Keywords: Homoiminosugar; Azasugar; Pyrrolidine; Azepane;
Epoxyamide; Epoxide opening.
With a view to broadening the utility of the epoxyamide
1, we have reanalysed our retrosynthetic route leading to
hydroxylated azepane derivatives 5 (Scheme 2). In a
* Corresponding author. Tel.: +34-952134260; fax: +34-952131941;
e-mail: pino@uma.es
z
Recently deceased. In memoriam.
0040-4039/$ - see front matter Ó 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2004.01.144

2612
C. Assiego et al. / Tetrahedron Letters 45 (2004) 2611–2613
(S) configuration at C-6.14 The complete connectivity of
1
the carbon and hydrogen atoms was ascertained by 2D
NMR experiments.
H
N
OR
CONEt
OR
2
OR´
TrO
Another structural confirmation of 3a was made with
the formation of the bicyclic compound 10 that could
only be formed by the syn disposition of the substituents
at C-3 and C-6 (Scheme 4). The chloride 9a was
obtained by reaction of 3a with mesyl chloride: the
initially formed mesylated derivative 8 was spontane-
ously transformed into the chloride 9a (13C NMR:
d 52.6 ppm, C-2). The introduction of the chlorine in 9a
was confirmed by high resolution X-ray photoelectron
spectroscopy. No rearrangement was observed15 from
the possible nitrogen-assisted retention of configuration
at C-2 (Scheme 4). Selective de-O-tritylation of 9a into
9b (80%) and subsequent displacement of the chloride in
basic medium afforded 10 in 66% yield. This transfor-
mation could be followed by 1H NMR using CD3OD as
the solvent (disappearance of the peaks 4.67 (d, H-2),
4.70 (t, H-5) and 4.86 (t, H-4) of product 9b and
appearance of 4.7–4.9 (m, H-4,5) of product 10). The
absolute configuration at C-2 could not be determined
from the NMR data.
R´
CONEt
2
R´´
N
O
O
3
O
O
5a R= TIPS, R´= OH, R´´= H
5b R´= H; R, R´´= OBn
6a R= TIPS, R´= H
6b R, R´= Bn
Scheme 2.
previous communication, we have mentioned the aze-
pane 5a as a side product in the formation of a piper-
idine derivative from precursor 6a.8 Now, we present an
alternative synthesis of azepane 5b as the sole product,
starting with the introduction of an azide at C-2 of 1,
followed by the protection of the free hydroxyls to give
6b.
For the synthesis of pyrrolidines, our strategy was based
on the introduction of an amino precursor at C-6,
according to the retrosynthetic analysis. In order to
avoid a nucleophilic attack of the epoxide, an oxidation
of the C-6 hydroxyl was envisaged to obtain a carbonyl
group which should compete to attract the nucleophile.
Indeed, the pyrrolidine 3a could be obtained in only two
steps from the epoxyamide 1 (Scheme 3), that is, five
steps from
D-ribose. Oxidation of the hydroxyl group
was performed with dimethyl sulfoxide and acetic
anhydride giving 7. Other methods employed (PCC or
Swern oxidations) gave lower yields. The ketone 7 was
condensed with benzyl amine and the resulting imine
reduced with NaBH3CN in one pot, giving the isomer 3a
through a regioselective and stereospecific intramolecu-
lar epoxide opening. It is to be noted that the nucleo-
philic attack at C-3 in the epoxyamide is due to a 5-exo
intramolecular cyclisation process. In the intermolecular
processes, we have always observed the nitrogen attack
at C-2.8;12 The high stereoselectivity observed in the
formation of 3a could be attributed to the approach of
the hydride from the less hindered face (b) in a favoured
conformation to start the cyclisation. In deprotected
substrates, it has been reported that the opposite stereo-
selectivity was observed in a similar reductive amination
reaction.13 The NMR data of 3a and its acetate 3b were
consistent with the presence of a pyrrolidine ring with
Bn
N
O
O
H
TrO
TrO
CONEt
2
CONEt
2
b
a
1
60%
OR
80%
O
O
O
O
7
3a R= H
3b R= Ac
c
Scheme 3. Reagents and conditions: (a) DMSO, Ac2O, 24 h; (b)
BnNH2, ZnCl2, NaBH3CN, EtOH, 7 h; (c) Ac2O, py, 48 h.
Scheme 4. Reagents and conditions: (a) MsCl, py, 24 h; (b) TFA,
CDCl3, 55 min; (c) CD3Ona/CD3OD, 72 h.

C. Assiego et al. / Tetrahedron Letters 45 (2004) 2611–2613
2613
References and notes
1. Wheatley, J. R.; Nash, R. J.; Watson, A. A.; Griffiths,
R. C.; Butters, T. D.; Muller, M.; Watkin, D. J.; Winkler,
D. A.; Fleet, G. W. J. J. Chem. Soc., Perkin Trans. 1 1999,
2735–2745.
2. Martin, O. R.; Saavedra, O. M.; Xie, F.; Liu, L.; Picasso,
S.; Vogel, P.; Kizu, H.; Asano, N. Bioorg. Med. Chem.
2001, 9, 1269–1278, and references cited therein.
3. Shilvock, J. P.; Fleet, G. W. J. Synlett 1998, 554–556.
4. Ikeda, K.; Takahashi, M.; Nishida, M.; Miyauchi, M.;
Kizu, H.; Kameda, Y.; Arisawa, M.; Watson, A. A.; Nash,
R. J.; Fleet, G. W. J.; Asano, N. Carbohydr. Res. 2000,
323, 73–80.
5. Asano, N.; Kato, A.; Miyauchi, M.; Kizu, H.; Kameda,
Y.; Watson, A. A.; Nash, R. J.; Fleet, G. W. J. J. Nat.
Prod. 1998, 61, 625–628.
6. Martin, O. R.; Compain, P.; Kizu, H.; Asano, N. Bioorg.
Med. Chem. Lett. 1999, 9, 3171–3174.
7. Asano, N.; Nishida, M.; Miyauchi, M.; Ikeda, K.;
Yamamoto, M.; Kizu, H.; Kameda, Y.; Watson, A. A.;
Nash, R. J.; Fleet, G. W. J. Phytochemistry 2000, 53, 379–
382.
8. Pino-Gonzalez, M. S.; Assiego, C.; Lopez-Herrera, F. J.
Tetrahedron Lett. 2003, 44, 8353–8356.
Scheme 5. Reagents and conditions: (a) NaN3, AcOH, DMF; (b)
BnBr, NaH, TBAI, THF, several days; (c) 5% TFA in CH2Cl2; (d)
MsCl, py, 0 °C, 15 h; (e) (i) Ph3P, CH2Cl2, 17 h, (ii) H2O, K2CO3, 48 h.
9. Synthesis of a,b-epoxyamides with sulfur ylides: (a)
ꢀ
Synthesis of 1: Lopez-Herrera, F. J.; Pino-Gonzalez,
M. S.; Sarabia Garcıa, F.; Heras Lopez, A.; Ortega-
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
Alcantara, J. J.; Pedraza-Cebrian, M. G. Tetrahedron:
Asymmetry 1996, 7, 2065–2071, corrected H NMR data
1
In order to synthesise azepane derivatives, the azido
group was stereoselectively introduced in 1 and the
hydroxyls benzylated giving, after 6 days at rt, a mixture
of 6b (31%) and monobenzylated product that could be
isolated and further rebenzylated. Hydrolysis of the
trityl group, mesylation, then reduction of the azido
group afforded the acyclic amine 12 and subsequent
cyclisation gave the azepane 5b. The structural assign-
ments were consistent with the NMR data (Scheme 5).16
ꢀ
in Ref. 8; (b) Lopez-Herrera, F. J.; Heras-Lopez, A. M.;
Pino Gonzalez, S.; Sarabia Garcıa, F. J. Org. Chem. 1996,
ꢀ
ꢀ
ꢀ
ꢀ
61, 8839–8848; (c) Lopez-Herrera, F. J.; Sarabia-Garcıa,
F.; Heras-Lopez, A.; Pino-Gonzalez, M. S. J. Org. Chem.
ꢀ
ꢀ
ꢀ
ꢀ
1998, 63, 9630–9634; (d) Lopez-Herrera, F. J.; Sarabia-
Garcıa, F.; Pedraza-Cebrian, G. M.; Pino-Gonzalez, M. S.
ꢀ
ꢀ
Tetrahedron Lett. 1999, 40, 1379–1380; (e) Valpuesta, M.;
ꢀ
Durante, P.; Lopez-Herrera, F. J. Tetrahedron 1993, 49,
9547–9560; (f) Izquierdo, I.; Plaza, M. T.; Robles, R.;
Mota, A. J. Tetrahedron: Asymmetry 2000, 11, 4509–4519.
10. Martin, O. R.; Liu, L.; Yang, F. Tetrahedron Lett. 1996,
37, 1991–1994.
The present syntheses of 3a, 3b and 5b from the
D-ribo
derivative 1 have demonstrated the utility of a,b-
epoxyamides in the formation of hydroxylated pyrrol-
idines and azepanes.17 To summarise, we have reported
an efficient methodology for the preparation of imino-
sugar derivatives with different ring sizes, combining
functional group transformations and regioselective
epoxide openings. Use of the above strategy for the
preparation of additional iminosugars, utilising different
monosaccharide starting materials, is under way in our
laboratory and will be reported in due course.
11. Yamashita, T.; Yasuda, K.; Kizu, H.; Kameda, Y.;
Watson, A. A.; Nash, R. J.; Fleet, G. W. J.; Asano, N.
J. Nat. Prod. 2002, 65, 1875–1881.
ꢀ
12. Valpuesta, M.; Durante, P.; Lopez-Herrera, F. J. Tetra-
hedron Lett. 1995, 36, 4681–4684.
13. Reitz, A. B.; Baxter, E. W. Tetrahedron Lett. 1990, 31,
6777–6780.
14. Selected data for 3a and 3b. Compound 3a. 1H NMR
(CDCl3): d 4.93 (dd, J5;6 ¼ 6:4, H-5), 4.70 (dd, J4;5 ¼ 5:2,
H-4), 4.57 (d, J2;3 ¼ 5:3, H-2), 2.75 (m, H-6), 2.68 (m,
J3;4 ¼ 6:3, H-3). 13C NMR (CDCl3): d 80.6 (C-5), 78.7 (C-
4), 67.7 (C-2), 67.6 (C-3), 66.4 (C-6). HRMS: calcd for
C40H47N2O5 (MHþ) 635.3485, found 635.3457. Com-
1
pound 3b. H NMR (CDCl3): d 5.51 (d, J2;3 ¼ 8:2, H-2),
4.72 (dd, J4;5 ¼ 6:0, H-4), 4.60 (dd, J5;6 ¼ 6:2, H-5), 3.06
(dd, J3;4 ¼ 4:6, H-3), 2.73 (m, H-6). 13C NMR (CDCl3): d
79.5 (C-5), 78.8 (C-4), 69.2 (C-2), 65.7 (C-3), 65.1 (C-6).
15. Cossy, J.; Dumas, C.; Gomez Pardo, D. Eur. J. Org.
Chem. 1999, 1693–1699.
Acknowledgements
This research was supported with funds from the
ꢀ
Direccion General de Investigacion Cientıfica y Tecnica
(Ref. BQU2001-1576) and from the Direccion General
ꢀ
ꢀ
ꢀ
16. Selected data for 5b. 1H NMR (CDCl3): d 4.50 (H-4, H-5),
4.05 (H-2), 3.50 (H-3), 3.20 (H-6), 3.10 (H-7,70). 13C NMR
(CDCl3): d 83.1 (C-3), 80.3 (C-6), 75.6 (C-4, C-5), 56.2 (C-
2), 44.5 (C-7).
17. Yields are given after product purification and are not
optimised. The deprotection of the products is under way
in our laboratory and will be reported in a full account of
our work.
ꢀ
ꢀ
ꢀ
de Universidades e Investigacion, Consejerıa de Edu-
cacion y Ciencia, Junta de Andalucıa (FQM 0158). We
ꢀ
ꢀ
ꢀ
wish to thank Dr. R. Rico (Universidad de Malaga) and
Dr. M. Angeles Pradera and Unidad de Espectroscopıa
ꢀ
de Masas de la Universidad de Sevilla for mass spectral
data.