Methyl 4-O-β-L-galactopyranosyl-β-D-glucopyranoside (methyl β-L-lactoside)

Article abstract of DOI:10.1107/S010827010504254X

Methyl β-l-lactoside, C₁₃H₂₄O₁₁, (II), is described by glycosidic torsion angles φ (O5_Gal - C1_Gal - O4_Glc - C4_Glc) and ψ (C1 _Gal - O1_Gal - C4_Glc - C5_Glc) of 93.89 (13) and -127.43 (13)°, respectively, where the ring atom numbering conforms to the convention in which C1 is the anomeric C atom and C6 is the exocyclic hydroxymethyl (CH₂OH) C atom in both residues (Gal is galactose and Glc is glucose). Substitution of l-Gal for d-Gal in the biologically relevant disaccharide, methyl β-lactoside [Stenutz, Shang & Serianni (1999). Acta Cryst. C55, 1719-1721], (I), significantly alters the glycosidic linkage interface. In the crystal structure of (I), one inter-residue (intramolecular) hydrogen bond is observed between atoms H3O _Glc and O5_Gal. In contrast, in the crystal structure of (II), inter-residue hydrogen bonds are observed between atoms H6O_Glc and O5_Gal, H6O_Glc and O6_Gal, and H3O _Glc and O2_Gal, with H6O_Glc serving as a donor with two intramolecular acceptors.

Full text of DOI:10.1107/S010827010504254X

organic compounds
Acta Crystallographica Section C
Crystal Structure
Communications
topology is to substitute an l sugar for a d sugar while
retaining the relative con®guration of the substituted residue.
This perturbation has been introduced into the structure of
ꢀ-d-Gal-(1!4)-ꢀ-d-GlcOCH3 (methyl ꢀ-lactoside; d-LAC),
(I), by substituting ꢀ-l-Gal at the ꢀ-d-Gal residue, yielding
ꢀ-l-Gal-(1!4)-ꢀ-d-GlcOCH3 (methyl ꢀ-l-lactoside; l-LAC),
(II).
ISSN 0108-2701
Methyl 4-O-b-L-galactopyranosyl-b-D-
glucopyranoside (methyl b-L-lactoside)
Qingfeng Pan, Bruce C. Noll and Anthony S. Serianni*
Department of Chemistry and Biochemistry, 251 Nieuwland Science Hall, University
of Notre Dame, IN 46556, USA
Correspondence e-mail: serianni.1@nd.edu
Received 21 November 2005
Accepted 20 December 2005
Online 21 January 2006
Methyl ꢀ-l-lactoside, C13H24O11, (II), is described by glyco-
sidic torsion angles ' (O5GalÐC1GalÐO4GlcÐC4Glc) and
(C1GalÐO1GalÐC4GlcÐC5Glc) of 93.89 (13) and 127.43 (13)ꢀ,
respectively, where the ring atom numbering conforms to the
convention in which C1 is the anomeric C atom and C6 is the
exocyclic hydroxymethyl (CH2OH) C atom in both residues
(Gal is galactose and Glc is glucose). Substitution of l-Gal for
d-Gal in the biologically relevant disaccharide, methyl
ꢀ-lactoside [Stenutz, Shang & Serianni (1999). Acta Cryst.
C55, 1719±1721], (I), signi®cantly alters the glycosidic linkage
interface. In the crystal structure of (I), one inter-residue
(intramolecular) hydrogen bond is observed between atoms
H3OGlc and O5Gal. In contrast, in the crystal structure of (II),
inter-residue hydrogen bonds are observed between atoms
Unlike d-LAC, l-LAC crystallizes in the solvent-free form
from methanol (Fig. 1). A comparison of representative
molecular parameters for (I) and (II) is shown in Table 2. The
bond lengths in (I) and (II) are, on average, similar, although
some signi®cant differences exist (e.g. C10ÐO50 and C2ÐO2).
The C1ÐO1ÐC40 bond angle is slightly smaller in l-LAC, and
this angle is larger than the C10Ð O10ÐCH3 glycosidic bond
angle in both structures.
H6OGlc and O5Gal, H6OGlc and O6Gal, and H3OGlc and O2Gal
,
with H6OGlc serving as a donor with two intramolecular
acceptors.
Comment
Structural studies of di- and oligosaccharides have focused to a
large extent on O-glycosidic linkages observed in nature. For
example, crystal structures of ꢀ-d-Gal-(1!4)-ꢁ- and ꢀ-d-
GlcOCH3, (I) (Stenutz et al., 1999; Pan et al., 2005), and
related ꢀ-(1!4) linkages, such as those found in ꢀ-d-Glc-
(1!4)-ꢀ-d-GlcOCH3 (Ham & Williams, 1970) and ꢀ-d-
GlcNAc-(1!4)-ꢀ-d-GlcNAc (Mo & Jensen, 1978), have been
investigated, and all contain linkage interfaces between
aldopyranosyl rings having the d con®guration (Gal is galac-
tose and Glc is glucose). As expected, these disaccharides
display common structural motifs dictated by their shared
linkage type. For example, ꢀ-(1!4) linkages involving
d sugars typically show inter-residue hydrogen bonds between
atoms H30O and O5, as shown in the ®rst scheme below for (I).
Thus, it may be inferred that efforts aimed at identifying the
various structural factors in¯uencing O-glycosidic linkage
conformations using natural systems will necessarily reveal
only a subset of the possible interface interactions. One way to
modulate the linkage interface and the overall molecular
The intra-ring torsion angles C1ÐC2ÐC3ÐC4 and C10Ð
C20ÐC30ÐC40 deviate signi®cantly from the idealized values
of 60ꢀ. This deviation is smaller for the intra-ring C1ÐO5Ð
C5ÐC4 and C10ÐO50ÐC50ÐC40 torsion angles, suggesting
that aldopyranosyl ring distortions result mainly from reduced
CÐCÐCÐC torsions. The glycosidic torsion angles ' are very
similar in (I) and (II), as indicated by the similar absolute
values of the C2ÐC1ÐO1ÐC40, O5ÐC1ÐO1ÐC40 and
H1ÐC1ÐO1ÐC40 torsion angles. In both structures, the '
o82 # 2006 International Union of Crystallography
DOI: 10.1107/S010827010504254X
Acta Cryst. (2006). C62, o82±o85

organic compounds
torsion angle is consistent with expectations based on
stereoelectronic factors (the exo-anomeric effect; C2 roughly
anti to C40) (Lemieux, 1971; Juaristi & Cuevas, 1995). In
contrast, the glycosidic torsion angle differs by ꢁ34ꢀ
between (I) and (II). Interestingly, in (II), assumes a nearly
eclipsed conformation [ 127.43 (13)ꢀ] in contrast to the more
nearly staggered conformation observed in (I) [ 161.3 (2)ꢀ],
indicating that signi®cant deviations from staggered confor-
mations are possible in glycosidic linkages. This ®nding
contrasts sharply with other mobile exocyclic bonds in
saccharides (e.g. C5ÐC6 bonds in aldohexopyranosyl rings)
where staggered or nearly staggered rotamers are energeti-
cally more stable than the eclipsed or near eclipsed rotamers
(Thibaudeau et al., 2004). Hydroxymethyl conformations in
the Gal (gauche±trans, gt) and Glc (gauche±gauche, gg) resi-
dues are the same in both structures.
the crystal structure. In l-LAC, atoms O60 and O5 are in
Ê
close proximity [3.0046 (15) A] and an inter-residue hydrogen
bond is observed between atoms H60O and O5 (see ®rst
scheme above). Furthermore, atom O60 is close to atom O6
0
Ê
Ê
[3.0817 (17) A], atom O3 is close to atom O2 [3.1357 (17) A],
and two additional hydrogen bonds are observed, with H60O
and H30O serving as donors. Thus, in (II), atom H60O parti-
cipates in two hydrogen bonds with atoms O5 and O6 (a three-
center hydrogen-bond system).
Recent NMR investigations of (I) indicate the presence of
hydrogen bonding between atoms H30O and O5 in a water/
acetone solvent at 253 K, based on the behavior of 3JHCOH and
3JCCOH values (Zhao et al., 2006). NMR studies of (I) in
aqueous solution indicate a mixture of gg and gt hydroxy-
methyl rotamers in the Glc residue based on an analysis of
3JHH values (Hayes et al., 1982). If hydrogen bonding between
atoms O60 and O5 of (II) exists in water/acetone solvent, not
only might the 3JHCOH and 3JCCOH values involving atom H60O
be affected, but also the distribution of gg and gt rotamers in
the Glc residue, with a shift towards the gg state expected.
Furthermore, comparisons of trans-glycoside J-couplings in d-
LAC and l-LAC may establish whether is shifted in solution
in a manner similar to that observed in the crystal. The results
of these NMR investigations will be reported elsewhere.
The crystal structure is a three-dimensional hydrogen-
bonded network composed of two-dimensional sheets that are
staggered to form columns in a three-dimensional lattice. In
the sheets, which are formed from a 6,3-network parallel to
(100), each molecule is connected to three additional mol-
ecules via two, two and three hydrogen bonds, respectively.
Additional hydrogen bonding connects each molecule to three
complementary molecules through two, one and one hydrogen
bond, extending the structure into alternating columns
parallel to the a axis. While there is no hydrogen bonding
between molecules of a given column, hydrogen bonds exist
The change in absolute con®guration of the Gal residue
affects the linkage interface signi®cantly. In d-LAC, atoms O30
Ê
and O5 are in close proximity (2.764 A) and an inter-residue
hydrogen bond is observed between atoms H30O and O5 in
Figure 1
The crystal structure of (II), showing the close proximity of atoms O60, O6
and O5, and atoms O30 and O2.
Figure 2
The hydrogen-bonding network generated from one molecule of (II).
Hydrogen bonds are shown as dashed lines and H atoms have been
omitted. The symmetry codes are as given in Table 1.
Figure 3
A view of the crystal packing of (II) along the a axis. Hydrogen bonds are
shown as dashed lines and H atoms have been omitted.
ꢂ
Acta Cryst. (2006). C62, o82±o85
Pan et al. C13H24O11 o83

organic compounds
Data collection
between molecules of neighboring columns. These alternating
1
columns are slipped with respect to each other by a trans-
Bruker X8-APEXII CCD
diffractometer
Rint = 0.031
2
ꢂmax = 31.5ꢀ
h = 6 ! 4
k = 16 ! 20
l = 30 ! 34
lation along the a axis. Details of the hydrogen bonding are
given in Table 1 and a representative view is shown in Fig. 2. A
view emphasizing the columnar packing is shown in Fig. 3.
Intermolecular hydrogen-bonding patterns observed for
(II) are very similar, but not identical, to those found for
natural disaccharides (Pan et al., 2005). Thus, atoms O10, O50
and O1 do not participate as hydrogen-bond acceptors. All
hydroxy H atoms are involved in hydrogen bonding, and most
hydroxy O atoms participate in either zero (O4) or one (O20,
O60, O2, O3 and O6) hydrogen bond as acceptors.
' and ! scans
13853 measured re¯ections
2964 independent re¯ections
2644 re¯ections with I > 2ꢄ(I)
Table 1
Hydrogen-bond geometry (A, ).
ꢀ
Ê
DÐHÁ Á ÁA
DÐH
HÁ Á ÁA
DÁ Á ÁA
DÐHÁ Á ÁA
i
O2ÐH2OÁ Á ÁO20
0.83 (3)
0.74 (3)
0.82 (2)
0.79 (3)
0.80 (3)
0.78 (3)
0.78 (3)
0.76 (3)
0.76 (3)
1.97 (3)
1.93 (3)
2.03 (2)
1.91 (3)
2.03 (3)
2.42 (2)
2.43 (3)
2.30 (3)
2.52 (2)
2.8042 (19)
2.6580 (19)
2.8475 (16)
2.6713 (19)
2.8263 (19)
3.1357 (17)
2.9719 (19)
3.0046 (15)
3.0817 (17)
175 (2)
167 (2)
174 (2)
163 (2)
169 (2)
153 (2)
128 (2)
157 (2)
133 (2)
O3ÐH3OÁ Á ÁO6ii
iii
The glycosidic linkage in (II) is a structural mimic of
linkages involving the biologically important ꢀ-l-fuco-
pyranosyl ring, (III), and the crystal structure of methyl ꢀ-l-
fucopyranosyl-(1!3)-ꢁ-d-glucopyranoside trihydrate has
O4ÐH4OÁ Á ÁO60
O6ÐH6OÁ Á ÁO3iv
O20ÐH20OÁ Á ÁO2i
O30ÐH30OÁ Á ÁO2
O30ÐH30OÁ Á ÁO30
O60ÐH60OÁ Á ÁO5
O60ÐH60OÁ Á ÁO6
i
È
È
been reported recently (Farnback et al., 2003). In di- and
oligosaccharides containing (III), however, potential inter-
residue hydrogen bonding is more restricted than for l-Gal,
since the l-Fuc ring lacks a hydroxy group at atom C6.
In addition to its modi®ed glycosidic linkage interface,
l-LAC displays an overall contact surface very different from
that of d-LAC. An identical but mirror-image contact surface
(and linkage interface) would be generated in the alternate
l-LAC structure composed of d-Gal and l-Glc residues.
Because both the linkage interface and surface topology are
affected by l-sugar substitution, this derivatization may prove
generally useful in the design of saccharide-based biological
agents or pharmaceuticals having novel binding properties or
functions.
Symmetry codes: (i) x ‡ 21; y ‡ 23; z; (ii) x ‡ 1; y ‡ 12; z ‡ 21; (iii) x ‡ 2, y ‡ 12,
z ‡ 12; (iv) x ‡ 2; y
;
z ‡ 12.
1
2
Table 2
Comparison of structural parameters in (I) and (II).
Parameter
(I)
(II)
Ê
Bond lengths (A)
C1ÐC2
1.527 (3)
1.516 (3)
1.425 (3)
1.413 (3)
1.387 (3)
1.384 (3)
1.437 (3)
1.414 (3)
1.423 (3)
1.426 (3)
1.418 (3)
1.511 (3)
1.508 (3)
2.764
1.528 (2)
1.534 (2)
1.4293 (17)
1.4352 (18)
1.3988 (18)
1.3821 (18)
1.4390 (17)
1.4313 (17)
1.4268 (18)
1.4392 (18)
1.4252 (17)
1.516 (2)
1.519 (2)
±
C10ÐC20
C1ÐO5
C10ÐO50
C1ÐO1
C10ÐO10
C40ÐO1
C2ÐO2
C4ÐO4
C6ÐO6
C20ÐO20
C5ÐC6
C50ÐC60
O30Á Á ÁO5
O(CH3OH)Á Á ÁO60
O60Á Á ÁO5
O60Á Á ÁO6
O30Á Á ÁO2
Experimental
Compound (II) was prepared by coupling 2,3,4,6-tetra-O-acetyl-ꢁ-l-
galactopyranosyl trichloroacetimidate, (IV) (600 mg, 1.21 mmol),
with methyl 2,3,6-tri-O-benzyl-ꢀ-d-glucopyranoside, (V) (500 mg,
1.08 mmol), in the presence of silver tri¯uoromethanesulfonate
(350 mg, 1.4 mmol) in dichloromethane (20 ml) at 253 K. When the
reaction was complete (2 h), the product was deprotected at room
temperature with 10% Pd/C (400 mg) in EtOAc (20 ml), followed
by addition of a 0.1 M methanol solution of NaOCH3 (15 ml).
Compound (IV) was prepared by standard methods (Schmidt et al.,
1984) and (V) was prepared by reacting methyl 2,3-di-O-benzyl-4,6-
O-benzylidene-ꢀ-d-glucopyranoside (450 mg, 0.97 mmol) in tetra-
hydrofuran (30 ml) with NaBH3CN (0.85 g, 13.7 mmol), immediately
followed by the addition of 1 N HCl in Et2O (14 ml). Puri®cation of
(II) was achieved by chromatography on silica gel using methanol/
dichloromethane (1:4 v/v) as the solvent. Crystals were grown from a
methanol solution by slow evaporation at 277 K.
2.727
±
±
±
3.0046 (15)
3.0817 (17)
3.1357 (17)
±
Bond angles (ꢀ)
C1ÐO1ÐC40
116.2 (2)
113.7 (2)
114.60 (12)
112.56 (13)
C10ÐO10ÐCH3
O30ÐH30OÁ Á ÁO5
OÐH(CH3OH)Á Á ÁO60
O60ÐH60OÁ Á ÁO5
O60ÐH60OÁ Á ÁO6
O30ÐH30OÁ Á ÁO2
140.9
164.4
±
±
±
±
±
157
133
153
Torsion angles (ꢀ)
C1ÐC2ÐC3ÐC4
C10ÐC20ÐC30ÐC40
54.8 (2)
44.2 (3)
65.0 (2)
67.6 (2)
153.8 (2)
164.2 (2)
78.4 (2)
161.3 (2)
88.4 (2)
77.4 (3)
31.9
53.35 (16)
52.15 (16)
62.68 (15)
66.04 (15)
146.19 (12)
167.74 (13)
111.14 (13)
127.43 (13)
93.89 (13)
73.32 (16)
25.6
C1ÐO5ÐC5ÐC4
C10ÐO50ÐC50ÐC40
C2ÐC1ÐO1ÐC40 (')
C20ÐC10ÐO10ÐCH3 ('0)
C1ÐO1ÐC40ÐC30 ( )
C1ÐO1ÐC40ÐC50 ( )
O5ÐC1ÐO1ÐC40 (')
O50ÐC10ÐO10ÐCH3('')
H1ÐC1ÐO1ÐC40 (')
C1ÐO1ÐC40ÐH40 ( )
O5ÐC5ÐC6ÐO6 (!)
O50ÐC50ÐC60ÐO60 (!0)
Crystal data
C13H24O11
Mr = 356.32
Orthorhombic, P212121
Ê
a = 4.7087 (4) A
b = 14.1428 (12) A
c = 23.2401 (19) A
Ê
V = 1547.7 (2) A
Mo Kꢁ radiation
Cell parameters from 4991
re¯ections
ꢂ = 2.9±31.3ꢀ
1
Ê
Ê
ꢃ = 0.14 mm
T = 100 (2) K
43.7
57.3 (2) (gt)
54.6 (2) (gg)
7.3
58.02 (16) (gt)
69.22 (16) (gg)
3
Needle, colorless
0.35 Â 0.07 Â 0.05 mm
Z = 4
Dx = 1.529 Mg m
3
ꢂ
o84 Pan et al. C13H24O11
Acta Cryst. (2006). C62, o82±o85

organic compounds
Re®nement
Supplementary data for this paper are available from the IUCr electronic
archives (Reference: JZ1778). Services for accessing these data are
described at the back of the journal.
Re®nement on F2
R[F2 > 2ꢄ(F2)] = 0.033
wR(F2) = 0.088
S = 1.04
2964 re¯ections
246 parameters
H atoms: see below
w = 1/[ꢄ2(Fo2) + (0.0559P)2
+ 0.1003P]
where P = (Fo2 + 2Fc2)/3
(Á/ꢄ)max = 0.002
3
Ê
References
Áꢅmax = 0.45 e A
3
Ê
0.21 e A
Áꢅmin
=
Allen, F. H., Johnson, O., Shields, G. P., Smith, B. R. & Towler, M. (2004). J.
Appl. Cryst. 37, 335±338.
Bruker (2004). APEX2 and SAINT. Bruker AXS Inc., Madison, Wisconsin,
USA.
È È
Farnback, M., Eriksson, L. & Widmalm, G. (2003). Acta Cryst. C59, o171±
o173.
Flack, H. D. (1983). Acta Cryst. A39, 876±881.
Ham, J. T. & Williams, D. G. (1970). Acta Cryst. B26, 1373±1383.
Hayes, M. L., Serianni, A. S. & Barker, R. (1982). Carbohydr. Res. 100, 87±
101.
Juaristi, E. & Cuevas, G. (1995). The Anomeric Effect. Boca Raton, Florida,
USA: CRC Press.
Lemieux, R. U. (1971). Pure Appl. Chem. 25, 527±548.
Mo, F. & Jensen, L. H. (1978). Acta Cryst. B34, 1562±1569.
Pan, Q., Noll, B. C. & Serianni, A. S. (2005). Acta Cryst. C61, o674±o677.
Schmidt, R. R., Michel, J. & Roose, M. (1984). Liebigs Ann. Chem. pp. 1343±
1357.
Sheldrick, G. M. (1998). XP. Bruker AXS Inc., Madison, Wisconsin,
USA.
Sheldrick, G. M. (2001). XCIF, XL and XS. Bruker AXS Inc., Madison,
Wisconsin, USA.
Hydroxy H atoms were located in a difference electron-density
map and freely re®ned in subsequent cycles of least-squares re®ne-
ment, including an isotropic displacement parameter. All other H
atoms were placed at calculated positions and allowed to ride on their
parent atoms. Displacement parameters for these H atoms were set at
1.2 times Ueq of the parent atom (1.5 times for methyl H atoms). 2143
Friedel pairs were merged in the ®nal stages of re®nement, and the
Flack (1983) parameter is thus meaningless. The absolute con®g-
uration was assumed from the synthesis.
Data collection: APEX2 (Bruker, 2004); cell re®nement: APEX2
and SAINT (Bruker, 2004); data reduction: SAINT and XPREP
(Sheldrick, 2003); program(s) used to solve structure: XS (Sheldrick,
2001); program(s) used to re®ne structure: XL (Sheldrick, 2001);
molecular graphics: XP (Sheldrick, 1998); software used to prepare
material for publication: XCIF (Sheldrick, 2001) and enCIFer (Allen
et al., 2004).
Sheldrick, G. M. (2003). XPREP. Bruker AXS Inc., Madison, Wisconsin,
USA.
Stenutz, R., Shang, M. & Serianni, A. S. (1999). Acta Cryst. C55, 1719±
1721.
The authors thank Professor Kenneth Henderson and Mr
Jacob Morris for useful discussions. The diffractometer was
acquired through generous support from the National Science
Foundation (award No. CHE-0443233).
Thibaudeau, C. Stenutz, R., Hertz, B., Klepach, T., Zhao, S., Wu, Q.,
Carmichael, I. & Serianni, A. S. (2004). J. Am. Chem. Soc. 126, 15668±
15685.
Zhao, H., Pan, Q., Zhang, W., Carmichael, I. & Serianni, A. S. (2006). J. Am.
Chem. Soc. Submitted.
ꢂ
Acta Cryst. (2006). C62, o82±o85
Pan et al. C13H24O11 o85