Two ethyl 2-deoxy-α-D-hexo-3,7-pyranoso-3-octulosonate derivatives

Article abstract of DOI:10.1107/S0108270101014974

Conformations of two ethyl 2-dexoy-α-D-hexo-3,7-pyranoso-3-octulosonate derivatives were studied. In each structure the anomeric hydroxy group formed an intramolecular hydrogen bond with the carbonyl O atom of the ethoxy-carbonylmethyl substituent. The co

Full text of DOI:10.1107/S0108270101014974

organic compounds
Acta Crystallographica Section C
Crystal Structure
Communications
ISSN 0108-2701
Two ethyl 2-deoxy-a-D-hexo-3,7-pyran-
oso-3-octulosonate derivatives
Anthony Linden,a* Xianfeng Lib and C. Kuan Leeb
aInstitute of Organic Chemistry, University of Zurich, Winterthurerstrasse 190,
È
CH-8057 Zurich, Switzerland, and bDepartment of Chemistry, National University
È
of Singapore, Kent Ridge, Singapore 119260
Correspondence e-mail: alinden@oci.unizh.ch
Figure 1
A view of the molecule of (I) showing the atom-labelling scheme.
Displacement ellipsoids are drawn at the 50% probability level and H
atoms are represented by spheres of arbitrary size.
Received 23 August 2001
Accepted 13 September 2001
In each of the two pyranoid sugars, ethyl 2-deoxy-4,5,6,8-tetra-
O-acetyl-ꢀ-d-gluco-3,7-pyranoso-3-octulosonate, C18H26O12,
and ethyl 2-deoxy-4,5,6,8-tetra-O-benzyl-ꢀ-d-galacto-3,7-pyran-
oso-3-octulosonate, C38H42O8, the anomeric con®guration is
ꢀ. The acetoxymethyl substituent on the hexopyranose ring of
the former compound and the ethoxycarbonylmethyl substi-
tuents in both sugars all have the gauche±trans conformation,
while the benzyloxymethyl substituent of the galactopyranose
sugar has the trans±gauche conformation. In each structure,
the anomeric hydroxy group forms an intramolecular
hydrogen bond with the carbonyl O atom of the ethoxy-
carbonylmethyl substituent.
d-galactose, respectively. Both sugars are ꢀ-anomers and the
4
pyranose ring in each compound has a slightly distorted C1
chair conformation. The ring puckering parameters (Cremer
Ê
Ê
& Pople, 1975) are Q = 0.570 (2) A, q2 = 0.062 (2) A, q3 =
ꢁ
ꢁ
Ê
0.567 (3) A, '2 = 68.5 (18) and ꢁ = 6.4 (2) for compound (I),
Ê
Ê
Ê
and Q = 0.590 (3) A, q2 = 0.054 (2) A, q3 = 0.588 (3) A, '2 =
157 (3)ꢁ and ꢁ = 5.2 (2)ꢁ for compound (II). The bond lengths
and angles exhibit normal values and generally agree with the
corresponding parameters found for other ꢀ-pyranose sugars
(Berman et al., 1967).
The conformation of the C5 acetoxymethyl group in (I) is
gauche±trans [O5ÐC5ÐC6ÐO6 72.1 (2)ꢁ and C4ÐC5ÐC6Ð
O6 ꢂ168.41 (17)ꢁ]. In contrast, the corresponding benzyl-
oxymethyl group in (II) has a trans±gauche conformation
[O5ÐC5ÐC6ÐO6 ꢂ170.90 (19)ꢁ and C4ÐC5ÐC6ÐO6
Comment
C-Glycosides are now widely used as chiral templates for the
synthesis of complex target molecules (Martin et al., 1991;
Jiang et al., 1996), many of which have been found to show
interesting and useful biological activities (Pougny et al., 1981;
Martin et al., 1991; Watson et al., 1994; Bichard et al., 1995).
One group of C-glycosides that have not been widely studied
for their enzyme-inhibition activities is that where the
compounds contain an exocyclic double bond at the anomeric
centre. We are presently interested in studying the synthesis
and structure of this class of derivatives and now report the
low-temperature crystal structures of ethyl 2-deoxy-4,5,-
6,8-tetra-O-acetyl-ꢀ-d-gluco-3,7-pyranoso-3-octulosonate, (I),
and ethyl 2-deoxy-4,5,6,8-tetra-O-benzyl-ꢀ-d-galacto-3,7-pyran-
oso-3-octulosonate, (II), which are precursors used in the
synthesis of their exocyclic alkenylic analogues.
Figure 2
Figs. 1 and 2 depict the correct absolute con®gurations of
compounds (I) and (II), which were assigned to agree with the
known chirality of the precursor sugars, viz. d-glucose and
A view of the molecule of (II) showing the atom-labelling scheme.
Displacement ellipsoids are drawn at the 50% probability level and H
atoms are represented by spheres of arbitrary size.
ꢀ
1418 # 2001 International Union of Crystallography
Printed in Great Britain ± all rights reserved
Acta Cryst. (2001). C57, 1418±1420

organic compounds
ꢂ50.0 (3)ꢁ]. In d-glucopyranose, a trans±gauche conformation
would be forbidden because of the resultant unfavourable
steric interaction between O4 and O6, but most galacto-
pyranoses have either the gauche±trans or the trans±gauche
conformation (Longchambon et al., 1975; Kanters et al., 1988),
with a slight preference for the gauche±trans form (Kanters et
al., 1978). The gauche±gauche conformation is not usually
observed for galactopyranoses, because this results in an
unfavourable 1,3-peri interaction between the synaxial atoms
O4 and O6.
The conformation of the C1 ethoxycarbonylmethyl substi-
tuent in each sugar is also gauche±trans [O5ÐC1ÐC9ÐC10
ꢂ66.7 (2)ꢁ and C2ÐC1ÐC9ÐC10 177.68 (18)ꢁ for (I), and
O5ÐC1ÐC14ÐC15 ꢂ62.5 (3)ꢁ and C2ÐC1ÐC14ÐC15
178.6 (2)ꢁ for (II)]. In each sugar, the anomeric C1 hydroxy
group forms an intramolecular hydrogen bond with the
ethoxycarbonyl O atom (Tables 1 and 2), thereby closing a six-
membered loop with a graph-set motif of S(6) (Bernstein et al.,
1995).
138.77 (Ar-C), 97.60 (C1), 80.34 (C3), 78.34 (C2), 74.77 (C4), 72.52,
73.29, 74.60, 75.15 (PhCH2), 70.08 (C5), 68.59 (C6 and C17), 40.56
(C14), 13.91 (C18). Analysis calculated for C38H42O8: C 72.86, H
6.71%; found: C 72.03, H 6.80%. Suitable crystals of each compound
were obtained by slow evaporation of their solutions in ethanol.
Compound (I)
Crystal data
C18H26O12
Dx = 1.328 Mg mꢂ3
Mo Kꢀ radiation
Cell parameters from 25
re¯ectionsꢁ
Mr = 434.39
Monoclinic, P21
a = 9.3164 (15) A
Ê
Ê
b = 13.038 (2) A
Ê
c = 9.3130 (15) A
ꢁ = 19.0±20.0
ꢄ = 0.11 mmꢂ1
ꢃ = 106.224 (12)ꢁ
T = 173 (1) K
Prism, colourless
0.46 Â 0.33 Â 0.24 mm
3
Ê
V = 1086.2 (3) A
Z = 2
Data collection
Rigaku AFC-5R diffractometer
!/2ꢁ scans
h = 0 ! 13
k = ꢂ18 ! 18
6676 measured re¯ections
3285 independent re¯ections
2717 re¯ections with I > 2ꢅ(I)
Rint = 0.024
l = ꢂ13 ! 12
3 standard re¯ections
every 150 re¯ections
intensity decay: none
ꢁ
max = 30ꢁ
Re®nement
Experimental
Re®nement on F2
R[F2 > 2ꢅ(F2)] = 0.038
wR(F2) = 0.106
S = 1.03
3285 re¯ections
280 parameters
H atoms treated by a mixture of
independent and constrained
re®nement
w = 1/[ꢅ2(Fo2) + (0.0554P)2
+ 0.1486P]
Compounds (I) and (II) were synthesized from the corresponding
2,3,4,6-tetra-O-benzyl-d-glucono- or -d-galactono-1,5-lactone using
the Reformatsky reaction (Shriner, 1942). The activated zinc used in
the reaction was prepared by washing zinc dust successively with 5%
HCl, distilled water, acetone, absolute ethanol and anhydrous ether,
and then drying at 373 K under vacuum. The Reformatsky reaction
produced the benzylated analogue of compound (I) and pure
compound (II), respectively. Compound (I) was then obtained by
converting its benzylated analogue to the acetylated derivative as
follows. A solution of ethyl 2-deoxy-4,5,6,8-tetra-O-benzyl-ꢀ-d-gluco-
3,7-pyranoso-3-octulosonate (0.5 g, 0.8 mmol) in ethanol (5 ml) was
hydrogenated [10% palladium-on-charcoal (80 mg) under hydrogen
(350 kPa)] for ca 12 h, after which thin-layer chromatography
(hexane/ethyl acetate, 3:1) showed that all the starting material had
been consumed. The solution was ®ltered, concentrated and acetyl-
ated with acetic anhydride (0.8 ml) in pyridine (1 ml). The reaction
mixture was worked up in the usual manner and puri®ed by ¯ash
column chromatography to give compound (I) (0.22 g, 64.2%; m.p.
383±385 K). Spectroscopic analysis: [ꢀ]D 40.5ꢁ (c 4.38, CHCl3);
1H NMR (CDCl3, ꢂ, p.p.m.): 5.52 (t, 1H, J3,4 = 9.5 Hz, H3), 5.09 (t, 1H,
J4,5 = 9.7 Hz, H4), 4.90 (d, 1H, J2,3 = 9.7 Hz, H2), 4.22±4.31 (m, 1H,
H5), 4.21 (q, 2H, J12,13 = 7.1 Hz, H12a,b), 2.58, 2.64 (2 Â s, 2H, H9a,b),
1.98, 2.01, 2.05, 2.09 (4 Â s, 12H, 4 Â CH3CO), 1.27 (t, 3H, H13a,b,c);
13C NMR (CDCl3, ꢂ, p.p.m.): 171.98, 170.01, 169.92, 169.52, 169.55
(C O), 96.12 (C1), 72.54 (C2), 70.92 (C3), 68.52, 68.01 (C4 and C5),
61.82, 61.50 (C1 and C6), 61.50 (C9), 39.62 (C12), 20.53, 20.59
(CH3CO), 13.88 (C13). Analysis calculated for C18H26O12: C 49.79, H
9.66%; found: C 49.52, H 9.61%. Compound (II) was obtained in
85.5% yield (m.p. 389±392 K). Spectroscopic analysis: [ꢀ]D ꢂ5.81 (c
10.69, CHCl3); 1H NMR (CDCl3, ꢂ, p.p.m.): 7.20±7.40 (m, 20H, Ar-H),
5.00, 4.95, 4.65, 4.61, 4.48, 4.42 (6 Â d, 6H, J = 11.4±11.8 Hz, 3 Â
PhCH2), 4.77 (s, 2H, H7a,b), 4.04±21 (m, 5H, H3±H5, H17a,b), 3.81
(d,1H,J2,3=9.8 Hz,H2),3.74(dd,1H,J5,6a=7.8 Hz,J6a,6b=9.3 Hz,H6a),
H6a), 3.48 (dd, 1H, H6b), 2.37, 2.83 (2 Â s, 2H, H14a,b), 1.27
(t, 3H, H18a,b,c); 13C NMR (CDCl3, ꢂ, p.p.m.): 172.44 (C O), 127.46,
127.58, 127.72, 128.02, 128.12, 128.28, 128.32, 128.47, 138.04, 138.38,
where P = (Fo2 + 2Fc2)/3
(Á/ꢅ)max = 0.001
Ê ꢂ3
Áꢆmax = 0.49 e A
Áꢆmin = ꢂ0.36 e A
Ê ꢂ3
Table 1
Hydrogen-bonding geometry (A, ) for (I).
ꢁ
Ê
DÐHÁ Á ÁA
DÐH
HÁ Á ÁA
DÁ Á ÁA
DÐHÁ Á ÁA
O1ÐH1Á Á ÁO10
0.79 (4)
2.11 (4)
2.794 (3)
146 (4)
Compound (II)
Crystal data
C38H42O8
Mr = 626.72
Mo Kꢀ radiation
Cell parameters from 25
re¯ectionsꢁ
Orthorhombic, P212121
Ê
a = 14.977 (2) A
Ê
b = 19.832 (7) A
ꢁ = 11.0±17.5
ꢄ = 0.09 mmꢂ1
Ê
c = 11.326 (2) A
Ê
V = 3364.0 (14) A
Z = 4
Dx = 1.237 Mg mꢂ3
T = 173 (1) K
Prism, colourless
0.50 Â 0.33 Â 0.30 mm
3
Data collection
Rigaku AFC-5R diffractometer
!/2ꢁ scans
h = 0 ! 19
k = ꢂ1 ! 25
4998 measured re¯ections
4302 independent re¯ections
3230 re¯ections with I > 2ꢅ(I)
Rint = 0.020
l = ꢂ1 ! 14
3 standard re¯ections
every 150 re¯ections
intensity decay: none
ꢁ
max = 27.5ꢁ
ꢀ
Acta Cryst. (2001). C57, 1418±1420
Linden, Li and Lee C18H26O12 and C38H42O8 1419

organic compounds
Re®nement
equivalents [3030 for (I) and 560 for (II)] were merged before the
®nal re®nements. The enantiomer used in each model was based on
the known chirality of the precursor sugars, viz. d-glucose and
d-galactose, from which (I) and (II), respectively, were synthesized.
For both compounds, data collection: MSC/AFC Diffractometer
Control Software (Molecular Structure Corporation, 1991); cell
re®nement: MSC/AFC Diffractometer Control Software; data reduc-
tion: TEXSAN (Molecular Structure Corporation, 1999); program(s)
used to solve structure: SHELXS97 (Sheldrick, 1997); program(s)
used to re®ne structure: SHELXL97 (Sheldrick, 1997); molecular
graphics: ORTEPII (Johnson, 1976); software used to prepare
material for publication: SHELXL97 and PLATON (Spek, 2001).
Re®nement on F2
R[F2 > 2ꢅ(F2)] = 0.042
wR(F2) = 0.103
S = 1.03
4302 re¯ections
421 parameters
H atoms treated by a mixture of
independent and constrained
re®nement
w = 1/[ꢅ2(Fo2) + (0.0379P)2
+ 0.5383P]
where P = (Fo2 + 2Fc2)/3
(Á/ꢅ)max = 0.001
Ê ꢂ3
Áꢆmax = 0.21 e A
Ê ꢂ3
Áꢆmin = ꢂ0.22 e A
Extinction correction: SHELXL97
(Sheldrick, 1997)
Extinction coef®cient: 0.0024 (5)
Table 2
Hydrogen-bonding geometry (A, ) for (II).
ꢁ
Ê
DÐHÁ Á ÁA
DÐH
HÁ Á ÁA
DÁ Á ÁA
DÐHÁ Á ÁA
Supplementary data for this paper are available from the IUCr electronic
archives (Reference: GG1083). Services for accessing these data are
described at the back of the journal.
O1ÐH1Á Á ÁO15
0.77 (3)
2.10 (3)
2.765 (3)
145 (3)
References
Examination of the structure of (I) with PLATON (Spek, 2001)
revealed that the unit-cell parameters can be transformed metrically
to an orthorhombic C lattice, but that the overall structure itself is not
consistent with the higher symmetry. For (I), the anisotropic dis-
placement ellipsoids for O16 and, to a lesser extent, O18 are signif-
icantly elongated. In addition, the maximum peak of residual electron
Berman, H. M., Chu, S. S. C. & Jeffrey, G. A. (1967). Science, 157, 1576±1577.
Bernstein, J., Davis, R. E., Shimoni, L. & Chang, N.-L. (1995). Angew Chem.
Int. Ed. Engl. 34, 1555±1573.
Bichard, C. J. F., Mitchell, E. P., Wormald, M. R., Watson, K. A., Johnson, L. N.,
Zographos, S. E., Koutra, D. D., Oikonomakos, N. G. & Fleet, G. W. J.
(1995). Tetrahedron Lett. 36, 2145±2148.
Cremer, D. & Pople, J. A. (1975). J. Am. Chem. Soc. 97, 1354±1358.
Flack, H. D. (1983). Acta Cryst. A39, 876±881.
Flack, H. D. & Bernardinelli, G. (2000). J. Appl. Cryst. 33, 1143±1148.
Jiang, S., Singh, G. & Wightman, R. H. (1996). Chem. Lett. pp. 67±68.
Johnson, C. K. (1976). ORTEPII. Report ORNL-5138. Oak Ridge National
Laboratory, Tennessee, USA.
Kanters, J. A., Scherrenberg, R. L., Lee¯ang, B. R., Kroon, J. & Mathlouthi, M.
(1988). Carbohydr. Res. 180, 175±182.
Kanters, K. A., Kock, A. J. H. M. & Roelofsen, G. (1978). Acta Cryst. B34,
3285±3288.
Longchambon, F., Ohannessian, J., Avenel, D. & Neuman, A. (1975). Acta
Cryst. B31, 2623±2627.
Martin, J. L., Veluraja, K., Ross, K., Johnson, L. N., Fleet, G. W. J., Ramsden,
N. G., Bruce, I., Orchard, M. G., Oikonomakos, N. G., Papageorgiou, A. C.,
Leonidas, D. D. & Tsitoura, H. S. (1991). Biochemistry, 30, 10101±10116.
Molecular Structure Corporation (1991). MSC/AFC Diffractometer Control
Software. MSC, 3200 Research Forest Drive, The Woodlands, TX 77381,
USA.
Molecular Structure Corporation (1999). TEXSAN. Version 1.10. MSC, 9009
New Trails Drive, The Woodlands, TX 77381±5209, USA.
Pougny, J.-R., Nassr, M. A. M. & SinayÈ, P. (1981). J. Chem. Soc. Chem.
Commun. pp. 375±376.
Ê ꢂ3
Ê
density of 0.49 e A is 1.0 A from O16 and the next highest peak is
Ê ꢂ3
only 0.25 e A . This suggests that these two atoms may be disor-
dered, particularly O16. Indeed, the position of O16 could be divided
Ê
into two almost equally occupied sites that are approximately 0.5 A
apart, and the re®nement of this model reduced R(F) to 0.035 and the
Ê ꢂ3
maximum peak of residual electron density to 0.25 e A . Never-
theless, the anisotropic displacement ellipsoid for one of the disor-
dered positions became even more elongated than that of O16 in the
ordered model, even when light restraints were applied, and more
severe restraints upset the realism of the geometric parameters.
Therefore, it was considered to be more appropriate to use the
ordered model in the ®nal re®nement. For each compound, the
methyl H atoms were constrained to an ideal geometry (CÐH =
Ê
0.98 A) with Uiso(H) = 1.5Ueq(C), but were allowed to rotate freely
about the CÐC bonds. The positions of the hydroxy H atoms were
re®ned freely, along with individual isotropic displacement para-
meters. All other H atoms were placed in geometrically idealized
Ê
positions (CÐH = 0.95±1.00 A) and constrained to ride on their
parent atoms, with Uiso(H) = 1.2Ueq(C). The absolute con®guration
could not be determined because of the absence of signi®cant
anomalous scatterers in the compound, and attempts to con®rm the
absolute structure by re®nement of the Flack parameter (Flack, 1983)
led to inconclusive values (Flack & Bernardinelli, 2000) for this
parameter [ꢂ0.7 (7) for (I) and ꢂ0.3 (11) for (II)]. Therefore, Friedel
È
Sheldrick, G. M. (1997). SHELXL97 and SHELXS97. University of Got-
tingen, Germany.
Shriner, R. L. (1942). Org. React. 1, 1±37.
Spek, A. L. (2001). PLATON. University of Utrecht, The Netherlands.
Watson, K. A., Mitchell, E. P., Johnson, L. N., Son, J. C., Bichard, C. J. F.,
Orchard, M. G., Fleet, G. W. J., Oikonomakos, N. G., Leonidas, D. D., Konyo,
M. & Papageorgiou, A. C. (1994). Biochemistry, 33, 5745±5748.
ꢀ
1420 Linden, Li and Lee
C18H26O12 and C38H42O8
Acta Cryst. (2001). C57, 1418±1420