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

Methyl α-lactoside, C₁₃H₂₄O₁₁, (I), is described by glycosidic torsion angles φ (O5_gal—C1_gal—O1_gal—C4_glc) and ψ (C1_gal—O1_gal—C4_glc—C5_glc), which have values of −93.52 (13) and −144.83 (11)°, 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. The linkage geometry is similar to that observed in methyl β-lactoside methanol solvate, (II), in which φ is −88.4 (4)° and ψ is −161.3 (4)°. As in (II), an inter­molecular O3_glc—H...O5_gal hydrogen bond is observed in (I). The hydroxy­methyl group conformation in both residues is gauche–trans, with torsion angles ω_gal (O5_gal—C5_gal—C6_gal—O6_gal) and ω_glc (O5_glc—C5_glc—C6_glc—O6_glc) of 69.15 (13) and 72.55 (14)°, respectively. The latter torsion angle differs substantially from that found for (II) [−54.6 (2)°; gauche–gauche]. Cocrystallization of methanol, which is hydrogen bonded to O6_glc in the crystal structure of (II), presumably affects the hydroxy­methyl conformation in the Glc residue in (II).

S1. Comment

The global structures of biologically important oligosaccharides are largely determined by the conformations of their constituent glycosidic linkages and of their exocyclic substituents, commonly hydroxymethyl and N-acetyl groups. Systematic structural comparisons within this series of biomolecule are hampered currently by the limited number of reported crystal structures containing biologically relevant glycosidic linkages in different structural contexts. Additional complications arise from the presence of either free or protected reducing ends, often in the latter case with diverse functionality.

The crystal structure of methyl α-lactoside, (I), has not been reported, but that of methyl β-lactoside methanol solvate, (2), has been published (Stenutz et al., 1999). Given the interest in deciphering the role of anomeric configuration in determining preferred linkage geometry, (I) was synthesized and crystallized from methanol. Unlike (II), (I) crystallizes without the inclusion of a solvent methanol molecule.

The structural parameters in (I) (Fig. 1) are compared with those observed in (II) and in methyl β- cellobioside methanol solvate, (III) (Ham and Williams, 1970), in Table 1. The C1'—O1' bond in (I) is considerably longer than in (II) and (III) (by ~0.02 Å), consistent with its change from an equatorial to an axial orientation. Likewise, the C4—O4 bonds in (I) and (II) are longer than that in (III) (by ~0.01 Å), again as a result of the axial orientation in the former. The C1'—O5' and C2'—O2' bond lengths also vary considerably in (I)–(III), with both bonds longer in (III) than in (I) and (II). Interestingly, these bond lengths differ in (II) and (III), despite their similar C1' configurations. Endocyclic C—C bond lengths in (I)–(III) involving non-anomeric C atoms appear on average to be slightly longer than the exocyclic C5—C6 bonds. This behavior may explain in part the generally larger ¹J_C5,C6 values observed in aldohexopyranosyl rings compared to the ¹J_CC values involving the ring C atoms (Wu et al., 1992).

The glycosidic C—O—C bond angles are larger for the internal linkages of (I)–(III) (~116°) than for terminal methyl aglycones (~113°) (Table 1), presumably owing to the greater steric demands of the reducing residue compared with the methyl group. Imaginary O3'—H···O5 bond angles vary from 134 to 150°, which is notably smaller than the 180° considered optimal for hydrogen bonding. Related bond angles for the methanol solvate hydrogen bonding in (II) and (III) are somewhat larger (~164°).

The internal pyranosyl ring torsion angles differ considerably from the idealized values of 60° (Table 1). Torsion angles solely involving C atoms appear more deformable than those involving the ring O atom, assuming values of between ~44 and 58°. In contrast, torsion angles involving the ring O atom are nearly ideal (~64°), although the values range from ~57 to 70°.

The internal glycosidic torsion angles in (II) more closely resemble those in (III) than those in (I). For example, using the heavy atoms as references, ϕ and ψ differ by ~2° in (II) and (III), whereas the corresponding values in (I) differ by ~4 (ϕ) and ~15° (ψ). In contrast, the absolute values of ϕ for the terminal glycosidic linkages differ by less than 4° in 1–3. Overall, however, glycosidic linkage conformations in 1–3 are highly conserved. Interestingly, recent NMR investigations of 1 and 2 in aqueous solution suggest highly similar conformations about their internal glycosidic linkages, implying that anomeric configuration at the reducing end of the disaccharide does not influence internal linkage conformation significantly.

The presence of methanol in the crystalline lattice influences the hydroxymethyl conformation significantly. In the solvates (II) and (III), methanol is hydrogen bonded to O6', leading to values of ω' of ~-55° (gg rotamer), whereas in 1, ω' assumes a value of 73° (gt rotamer). Values of ω are similar in 1–3 (gt rotamer) regardless of the configuration at atom C4. In contrast, recent NMR studies showed a roughly equal distribution of gg and gt rotamers in Glc monomers and a highly preferred gt rotamer, ~70%, in Gal monomers in aqueous solution (Thibaudeau et al., 2004).

Intramolecular hydrogen bonding between atoms O3' and O5 is observed in (I)–(III), with internuclear distances between the heavy atoms of ~2.8 Å (Table 1). This distance is comparable to that observed between the methanol O atom and atom O6' in (II) and (III), which averages 2.7 Å.

In the crystal structure of (I) (Fig. 2), intermolecular hydrogen bonds can be divided into two groups, viz. an infinite chain with hydrogen-bonds alternating between atoms O6 and O3, and a six-membered chain starting at atom HO4 through HO3' and ending at atom O5. For (II) and (III), a similar infinite chain is observed, but the finite chain contains seven members, including a link through a methanol molecule. This seven-membered chain starts at atom HO4 and ends at atom O5.

All hydroxy H atoms in (I) are involved in intermolecular hydrogen bonding as hydrogen donors, whereas the O atoms serve as acceptors to a maximum of one hydrogen bond. In this way, a two-dimensional hydrogen-bonded network in the (001) plane is developed. Atoms O1', O5', O1 and O4 are not involved in intermolecular hydrogen bonding, and atom O5 experiences intramolecular hydrogen bonding to O3' (see above). Similar behavior is observed in (II) and (III).

Solution structures of α-lactose have been reported as a monohydrate, (IV) (Fries et al., 1971), and as calcium complexes (Cook & Bugg, 1973; Bugg, 1973). The values of ϕ and ω in (IV) are very similar to those observed in (I) (Table 2). The pyranosyl ring distortions observed in (I) are also observed in (IV), and gt hydroxymethyl rotamers are found in (IV) as in (I) [the water molecule in the crystal structure of (IV) is not bonded to O6']. Interestingly, the C1—O1—C4' bond angle is considerably larger in (IV) (117°) than in (I) [115.26 (10)°]. The C1'—O1' bond is shorter in (IV) (1.387 Å) than in (I) [1.4012 19) Å], whereas the C1'—O5' bond is longer in (IV) than in (I). These bond-length differences are attributed in part to the changes in O1' substitution. Some of these trends are maintained when comparing (II) to β-lactose (V) (Hirotsu & Shimada, 1974) (Table 2). In this case, the C1—O1—C4' bond angles and C1'—O1' bond lengths are very similar, but the C1'—O5' bond is shorter in the glycoside. These results suggest that the effects of O1' substitution on proximal bond lengths and angles depend on anomeric configuration. Hydroxymethyl conformation in the Glc residue of (V) is gt, in contrast to the gg form observed in (II).

A comparison of the packing structure of the methanol solvates (II) and (III) with that of (I) is informative. In (II) and (III), the methyl aglycones are aligned along a well defined, relatively hydrophobic channel that includes the methanol solvent. This alignment appears critical to crystal formation in the methyl glycosides, as evidenced by the fact that a similar alignment is observed in (I), although apparently the structural requirements (i.e. hydrogen bonding) are met in this case without the need for cocrystallization of solvent methanol. Structures either lacking the methyl aglycone [e.g. (V)] or cocrystallizing with water [e.g. (IV)] do not appear to crystallize with similar channels.

Differences in packing structures, and thus hydrogen-bonding patterns, are likely to modulate the crystal structures of saccharides, thus complicating comparisons between apparently similar structures such as (I) and (IV). The presence of inter- and intramolecular hydrogen bonding, and packing forces, is expected to modulate bond lengths, angles and torsion angles in a fashion that is not currently predictable or quantifiable, and conclusions based on structural comparisons within this restricted set of structures must be viewed with this limitation in mind.

S2. Experimental

Compound (I) was prepared by coupling 2,3,4,6-tetra-O-acetyl-α-D-galactopyranosyl trichloroacetimidate with methyl 2,3,6-tri-O-benzyl-β-D-glucopyranoside in the presence of trimethylsilyl trifluoromethanesulfonate catalyst, followed by deprotection with palladium over carbon and sodium methoxide. The imidate donor was prepared by standard methods (Schmidt et al., 1984) and the acceptor was prepared from methyl 2,3-di-O-benzyl-4,6-O-benzylidene-β-D-glucopyranoside. Purification of (I) was achieved by chromatography on silica gel using methanol/dichloromethane (1:4) as the solvent. Crystals were grown from methanol by slow evaporation at 277 K. [Please give quantities of reagents and reaction conditions.]

S3. Refinement

All H atoms were clearly resolved in difference maps and were subsequently allowed for as riding, with C—H distances in the range 0.98–1.00 Å and O—H distances of 0.84 Å. For methyl and hydroxy H atoms, U_iso(H) was set at 1.5U_eq(C,O). For all other C atoms, U_iso(H) was set at 1.2U_eq(C).

Computing details

Data collection: APEX2 (Bruker–Nonius, 2004); cell refinement: APEX2 and SAINT (Bruker–Nonius, 2004); data reduction: SAINT and XPREP (Sheldrick, 2003); program(s) used to solve structure: XS (Sheldrick, 2001); program(s) used to refine 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).

Figures

	Fig. 1. The structure and atomic numbering of methyl α-lactoside. 50% probability displacement ellipsoids are shown for the C and O atoms. H atoms are shown as spheres of arbitrary radii.
	Fig. 2. The hydrogen-bonding network in methyl α-lactoside. [Symmetry codes: (i) x + 1/2, y − 1/2, z; (ii) −x + 3/2, y − 1/2, −z; (iii) x − 1/2, y − 1/2, z.]

methyl 4-O-β-D-galactopyranosyl-α-D-glucopyranoside

Crystal data

C13H24O11	F(000) = 760
Mr = 356.32	Dx = 1.533 Mg m^−3
Monoclinic, C2	Mo Kα radiation, λ = 0.71073 Å
Hall symbol: C 2y	Cell parameters from 8868 reflections
a = 14.6984 (3) Å	θ = 2.8–31.5°
b = 5.0061 (1) Å	µ = 0.14 mm^−1
c = 21.1204 (5) Å	T = 100 K
β = 96.5502 (12)°	Needle, colourless
V = 1543.93 (6) Å^3	0.36 × 0.22 × 0.18 mm
Z = 4

Data collection

Bruker SMART APEX CCD diffractometer	2758 reflections with I > 2σ(I)
Radiation source: fine-focus sealed tube, Siemens KFFMO2K-90	Rint = 0.027
Graphite monochromator	θmax = 31.5°, θmin = 1.0°
Detector resolution: 8.33 pixels mm^-1	h = −21→19
ϕ and ω scans	k = −7→5
11448 measured reflections	l = −31→29
2845 independent reflections

Refinement

Refinement on F^2	Secondary atom site location: difference Fourier map
Least-squares matrix: full	Hydrogen site location: inferred from neighbouring sites
R[F^2 > 2σ(F^2)] = 0.032	H-atom parameters constrained
wR(F^2) = 0.095	w = 1/[σ^2(Fo^2) + (0.0697P)^2 + 0.4616P] where P = (Fo^2 + 2Fc^2)/3
S = 1.04	(Δ/σ)max = 0.043
2845 reflections	Δρmax = 0.55 e Å^−3
225 parameters	Δρmin = −0.24 e Å^−3
1 restraint	Absolute structure: Flack (1983)
Primary atom site location: structure-invariant direct methods	Absolute structure parameter: 0.2 (4)

Special details

Geometry. All e.s.d.'s (except the e.s.d. in the dihedral angle between two l.s. planes) are estimated using the full covariance matrix. The cell e.s.d.'s are taken into account individually in the estimation of e.s.d.'s in distances, angles and torsion angles; correlations between e.s.d.'s in cell parameters are only used when they are defined by crystal symmetry. An approximate (isotropic) treatment of cell e.s.d.'s is used for estimating e.s.d.'s involving l.s. planes.

Refinement. The structure was solved from a partial data set while data collection was in progress using dual-space methods available as XM in the SHELXTL (Bruker, 2004) program package. This revealed the complete non-hydrogen structure where direct methods failed. After making proper atom assignments, the structure was refined against the complete 4-fold redundant data by usual least-squares techniques. H atoms were placed at calculated geometries and allowed to ride on the position of the parent atom. Hydrogen thermal parameters were set to 1.2× the equivalent isotropic U of the parent atom, 1.5× for methyl H atoms and H atoms attached to oxygen.

The numbering scheme used is usual and customary for the carbohydrate community.

The largest peak in the final difference map, 0.548 e⁻/ų, is located on the midpoint of the C1—O1 bond. Eight of the 9 remaining large peaks are similarly located along bonds. The 9t h peak, 0.33 e⁻/ų, appears as a possible alternate position for the hydrogen atom bound to O3. No attempt was made to model this position.

The hydrogen bonding network may be derived from the following bonds and angles:

Hydrogen bonds with H.·A < r(A) + 2.000 Angstroms and <DHA > 110 °.

D—H d(D—H) d(H.·A) <DHA d(D.·A) A

O2—H2A 0.840 1.996 160.62 2.803 O3' [x + 1/2, y − 1/2, z]

O3—H3A 0.840 1.917 169.29 2.747 O6 [x + 1/2, y − 1/2, z]

O4—H4A 0.840 1.985 178.42 2.824 O3 [−x + 3/2, y − 1/2, −z]

O6—H6 0.840 1.837 166.54 2.662 O2 [x − 1/2, y − 1/2, z]

O2'-H2'A 0.840 1.894 159.34 2.696 O6' [x − 1/2, y − 1/2, z]

O6'-H6' 0.840 1.939 152.88 2.714 O2' [x + 1/2, y − 1/2, z]

Refinement of F² against ALL reflections. The weighted R-factor wR and goodness of fit S are based on F², conventional R-factors R are based on F, with F set to zero for negative F². The threshold expression of F² > σ(F²) is used only for calculating R-factors(gt) etc. and is not relevant to the choice of reflections for refinement. R-factors based on F² are statistically about twice as large as those based on F, and R- factors based on ALL data will be even larger.

Fractional atomic coordinates and isotropic or equivalent isotropic displacement parameters (Ų)

	x	y	z	Uiso*/Ueq
C1	0.69499 (8)	−0.0078 (3)	0.18864 (6)	0.0097 (2)
H1	0.6897	0.1776	0.1709	0.012*
C2	0.77456 (8)	−0.1537 (3)	0.16277 (6)	0.0098 (2)
H2	0.7762	−0.3456	0.1760	0.012*
C3	0.75616 (8)	−0.1286 (3)	0.08958 (6)	0.0096 (2)
H3	0.7533	0.0668	0.0797	0.012*
C4	0.66199 (9)	−0.2462 (3)	0.06414 (6)	0.0099 (2)
H4	0.6477	−0.1926	0.0185	0.012*
C5	0.58682 (8)	−0.1354 (3)	0.10148 (6)	0.0101 (2)
H5	0.5754	0.0560	0.0896	0.012*
C6	0.49729 (9)	−0.2886 (3)	0.09011 (6)	0.0135 (2)
H6A	0.4798	−0.3125	0.0438	0.016*
H6B	0.5045	−0.4674	0.1101	0.016*
C1'	0.61032 (9)	0.4679 (3)	0.39531 (6)	0.0122 (2)
H1'	0.5912	0.6283	0.4188	0.015*
C2'	0.54148 (9)	0.4242 (3)	0.33643 (6)	0.0111 (2)
H2'	0.5422	0.5877	0.3093	0.013*
C3'	0.57005 (9)	0.1893 (3)	0.29744 (6)	0.0106 (2)
H3'	0.5680	0.0213	0.3229	0.013*
C4'	0.66794 (9)	0.2367 (3)	0.28202 (6)	0.0100 (2)
H4'	0.6683	0.3889	0.2514	0.012*
C5'	0.73327 (9)	0.2988 (3)	0.34186 (6)	0.0107 (2)
H5'	0.7423	0.1358	0.3692	0.013*
C6'	0.82460 (9)	0.3937 (3)	0.32343 (6)	0.0131 (2)
H6'A	0.8443	0.2735	0.2903	0.016*
H6'B	0.8173	0.5753	0.3050	0.016*
C7'	0.64991 (14)	0.2851 (5)	0.49772 (7)	0.0302 (4)
H7'A	0.7156	0.3133	0.4964	0.045*
H7'B	0.6406	0.1282	0.5240	0.045*
H7'C	0.6234	0.4427	0.5161	0.045*
O1	0.70545 (6)	0.0019 (2)	0.25464 (4)	0.01065 (18)
O2	0.85833 (6)	−0.0244 (2)	0.18499 (5)	0.01283 (19)
H2A	0.8929	−0.1323	0.2066	0.019*
O3	0.82630 (7)	−0.2370 (2)	0.05616 (4)	0.01276 (19)
H3A	0.8509	−0.3669	0.0767	0.019*
O4	0.66407 (7)	−0.5309 (2)	0.06738 (5)	0.0138 (2)
H4A	0.6660	−0.5939	0.0307	0.021*
O5	0.61288 (6)	−0.1533 (2)	0.16915 (4)	0.01033 (19)
O6	0.42756 (7)	−0.1416 (2)	0.11727 (5)	0.0154 (2)
H6	0.4012	−0.2424	0.1412	0.023*
O1'	0.60613 (8)	0.2429 (3)	0.43431 (5)	0.0169 (2)
O2'	0.45171 (7)	0.3967 (2)	0.35413 (5)	0.0145 (2)
H2'A	0.4450	0.2419	0.3682	0.022*
O3'	0.50573 (7)	0.1704 (3)	0.24129 (5)	0.0160 (2)
H3'A	0.5200	0.0416	0.2189	0.024*
O5'	0.69893 (7)	0.5137 (2)	0.37744 (5)	0.01307 (19)
O6'	0.89376 (7)	0.3986 (2)	0.37696 (5)	0.0145 (2)
H6'	0.9160	0.2449	0.3830	0.022*

Atomic displacement parameters (Ų)

	U^11	U^22	U^33	U^12	U^13	U^23
C1	0.0064 (5)	0.0112 (6)	0.0115 (5)	−0.0009 (4)	0.0006 (4)	−0.0002 (4)
C2	0.0065 (5)	0.0110 (6)	0.0119 (5)	−0.0001 (4)	0.0014 (4)	0.0004 (4)
C3	0.0070 (5)	0.0118 (6)	0.0103 (5)	0.0010 (4)	0.0023 (4)	0.0001 (4)
C4	0.0082 (5)	0.0105 (5)	0.0111 (5)	−0.0003 (4)	0.0018 (4)	−0.0002 (4)
C5	0.0075 (5)	0.0119 (6)	0.0108 (5)	−0.0006 (4)	0.0003 (4)	0.0003 (4)
C6	0.0072 (5)	0.0193 (7)	0.0142 (5)	−0.0025 (5)	0.0021 (4)	−0.0022 (5)
C1'	0.0098 (5)	0.0137 (6)	0.0135 (5)	0.0000 (5)	0.0029 (4)	−0.0015 (4)
C2'	0.0083 (5)	0.0108 (6)	0.0145 (5)	0.0007 (4)	0.0024 (4)	−0.0020 (4)
C3'	0.0072 (5)	0.0107 (5)	0.0135 (5)	−0.0001 (4)	0.0004 (4)	−0.0016 (4)
C4'	0.0079 (5)	0.0098 (6)	0.0122 (5)	0.0004 (4)	0.0011 (4)	−0.0018 (4)
C5'	0.0081 (5)	0.0123 (6)	0.0117 (5)	−0.0011 (5)	0.0008 (4)	−0.0023 (4)
C6'	0.0080 (5)	0.0168 (6)	0.0144 (5)	−0.0023 (5)	0.0008 (4)	−0.0020 (5)
C7'	0.0353 (9)	0.0399 (11)	0.0146 (6)	0.0025 (9)	0.0001 (6)	0.0046 (7)
O1	0.0096 (4)	0.0117 (4)	0.0107 (4)	0.0015 (4)	0.0012 (3)	−0.0012 (4)
O2	0.0059 (4)	0.0165 (5)	0.0156 (4)	−0.0018 (4)	−0.0007 (3)	0.0017 (4)
O3	0.0100 (4)	0.0159 (5)	0.0133 (4)	0.0044 (4)	0.0055 (3)	0.0021 (4)
O4	0.0176 (5)	0.0099 (4)	0.0144 (4)	−0.0006 (4)	0.0042 (3)	−0.0015 (4)
O5	0.0065 (4)	0.0141 (5)	0.0105 (4)	−0.0022 (3)	0.0013 (3)	−0.0012 (3)
O6	0.0074 (4)	0.0176 (5)	0.0218 (5)	0.0019 (4)	0.0050 (3)	0.0041 (4)
O1'	0.0159 (5)	0.0208 (6)	0.0144 (4)	0.0003 (4)	0.0030 (3)	0.0036 (4)
O2'	0.0082 (4)	0.0134 (5)	0.0226 (5)	0.0011 (4)	0.0050 (3)	−0.0020 (4)
O3'	0.0089 (4)	0.0210 (6)	0.0172 (4)	0.0023 (4)	−0.0020 (3)	−0.0073 (4)
O5'	0.0095 (4)	0.0148 (5)	0.0153 (4)	−0.0023 (4)	0.0032 (3)	−0.0048 (4)
O6'	0.0089 (4)	0.0144 (5)	0.0193 (4)	0.0002 (4)	−0.0021 (3)	−0.0038 (4)

Geometric parameters (Å, º)

C1—O1	1.3857 (14)	C2'—H2'	1.0000
C1—O5	1.4295 (16)	C3'—O3'	1.4329 (16)
C1—C2	1.5316 (17)	C3'—C4'	1.5295 (17)
C1—H1	1.0000	C3'—H3'	1.0000
C2—O2	1.4225 (16)	C4'—O1	1.4467 (17)
C2—C3	1.5437 (17)	C4'—C5'	1.5296 (18)
C2—H2	1.0000	C4'—H4'	1.0000
C3—O3	1.4219 (15)	C5'—O5'	1.4364 (17)
C3—C4	1.5432 (19)	C5'—C6'	1.5159 (18)
C3—H3	1.0000	C5'—H5'	1.0000
C4—O4	1.4273 (18)	C6'—O6'	1.4317 (16)
C4—C5	1.5325 (18)	C6'—H6'A	0.9900
C4—H4	1.0000	C6'—H6'B	0.9900
C5—O5	1.4397 (15)	C7'—O1'	1.434 (2)
C5—C6	1.5183 (18)	C7'—H7'A	0.9800
C5—H5	1.0000	C7'—H7'B	0.9800
C6—O6	1.4333 (17)	C7'—H7'C	0.9800
C6—H6A	0.9900	O2—H2A	0.8400
C6—H6B	0.9900	O3—H3A	0.8400
C1'—O1'	1.4012 (19)	O4—H4A	0.8400
C1'—O5'	1.4156 (15)	O6—H6	0.8400
C1'—C2'	1.5268 (19)	O2'—H2'A	0.8400
C1'—H1'	1.0000	O3'—H3'A	0.8400
C2'—O2'	1.4185 (15)	O6'—H6'	0.8400
C2'—C3'	1.5221 (19)
O1—C1—O5	107.49 (9)	O2'—C2'—H2'	107.5
O1—C1—C2	111.95 (10)	C3'—C2'—H2'	107.5
O5—C1—C2	107.95 (11)	C1'—C2'—H2'	107.5
O1—C1—H1	109.8	O3'—C3'—C2'	107.58 (11)
O5—C1—H1	109.8	O3'—C3'—C4'	112.45 (10)
C2—C1—H1	109.8	C2'—C3'—C4'	108.64 (11)
O2—C2—C1	109.44 (11)	O3'—C3'—H3'	109.4
O2—C2—C3	109.85 (10)	C2'—C3'—H3'	109.4
C1—C2—C3	105.35 (10)	C4'—C3'—H3'	109.4
O2—C2—H2	110.7	O1—C4'—C3'	111.70 (11)
C1—C2—H2	110.7	O1—C4'—C5'	105.28 (10)
C3—C2—H2	110.7	C3'—C4'—C5'	111.88 (10)
O3—C3—C4	110.72 (11)	O1—C4'—H4'	109.3
O3—C3—C2	114.33 (10)	C3'—C4'—H4'	109.3
C4—C3—C2	111.43 (10)	C5'—C4'—H4'	109.3
O3—C3—H3	106.6	O5'—C5'—C6'	105.90 (11)
C4—C3—H3	106.6	O5'—C5'—C4'	111.08 (11)
C2—C3—H3	106.6	C6'—C5'—C4'	110.02 (10)
O4—C4—C5	110.45 (11)	O5'—C5'—H5'	109.9
O4—C4—C3	110.53 (12)	C6'—C5'—H5'	109.9
C5—C4—C3	110.70 (11)	C4'—C5'—H5'	109.9
O4—C4—H4	108.4	O6'—C6'—C5'	111.86 (11)
C5—C4—H4	108.4	O6'—C6'—H6'A	109.2
C3—C4—H4	108.4	C5'—C6'—H6'A	109.2
O5—C5—C6	104.66 (10)	O6'—C6'—H6'B	109.2
O5—C5—C4	111.26 (10)	C5'—C6'—H6'B	109.2
C6—C5—C4	113.50 (11)	H6'A—C6'—H6'B	107.9
O5—C5—H5	109.1	O1'—C7'—H7'A	109.5
C6—C5—H5	109.1	O1'—C7'—H7'B	109.5
C4—C5—H5	109.1	H7'A—C7'—H7'B	109.5
O6—C6—C5	108.91 (12)	O1'—C7'—H7'C	109.5
O6—C6—H6A	109.9	H7'A—C7'—H7'C	109.5
C5—C6—H6A	109.9	H7'B—C7'—H7'C	109.5
O6—C6—H6B	109.9	C1—O1—C4'	115.26 (10)
C5—C6—H6B	109.9	C2—O2—H2A	109.5
H6A—C6—H6B	108.3	C3—O3—H3A	109.5
O1'—C1'—O5'	112.98 (12)	C4—O4—H4A	109.5
O1'—C1'—C2'	107.10 (12)	C1—O5—C5	111.98 (9)
O5'—C1'—C2'	110.51 (10)	C6—O6—H6	109.5
O1'—C1'—H1'	108.7	C1'—O1'—C7'	112.66 (14)
O5'—C1'—H1'	108.7	C2'—O2'—H2'A	109.5
C2'—C1'—H1'	108.7	C3'—O3'—H3'A	109.5
O2'—C2'—C3'	112.95 (11)	C1'—O5'—C5'	114.03 (11)
O2'—C2'—C1'	110.44 (10)	C6'—O6'—H6'	109.5
C3'—C2'—C1'	110.70 (11)

Hydrogen-bond geometry (Å, º)

D—H···A	D—H	H···A	D···A	D—H···A
O2—H2A···O3′^i	0.84	2.00	2.8026 (15)	161
O3—H3A···O6^i	0.84	1.92	2.7469 (14)	169
O4—H4A···O3^ii	0.84	1.99	2.8243 (14)	178
O6—H6···O2^iii	0.84	1.84	2.6619 (14)	167
O2′—H2′A···O6′^iii	0.84	1.89	2.6962 (14)	159
O6′—H6′···O2′^i	0.84	1.94	2.7138 (14)	153

Symmetry codes: (i) x+1/2, y−1/2, z; (ii) −x+3/2, y−1/2, −z; (iii) x−1/2, y−1/2, z.