3-De­oxy-[beta]-D-ribo-hexopyran­ose (3-de­oxy-[beta]-D-glucopyran­ose)

The [beta]-pyran­ose form, (III), of 3-deoxy-D-ribo-hexose (3-deoxy-D-glucose), C₆H₁₂O₅, crystallizes from water at 298 K in a slightly distorted ⁴C₁ chair conformation. Structural analyses of (III), [beta]-D-glucopyran­ose, (IV), and 2-deoxy-[beta]-D-arabino-hexopyran­ose (2-deoxy-[beta]-D-glucopyran­ose), (V), show significantly different C-O bond torsions involving the anomeric carbon, with the H-C-O-H torsion angle approaching an eclipsed conformation in (III) (-10.9°) compared with 32.8 and 32.5° in (IV) and (V), respectively. Ring carbon deoxy­genation significantly affects the endo- and exocyclic C-C and C-O bond lengths throughout the pyran­ose ring, with longer bonds generally observed in the monodeoxy­genated species (III) and (V) compared with (IV). These structural changes are attributed to differences in exocyclic C-O bond conformations and/or hydrogen-bonding patterns superimposed on the direct (intrinsic) effect of monodeoxy­genation. The exocyclic hydroxy­methyl conformation in (III) (gt) differs from that observed in (IV) and (V) (gg).

S1. Comment

As part of ongoing NMR investigations of non-enzymic protein glycation (Voziyan et al., 2003), ¹³C-labeled 3-deoxy-D-erythro-hexos-2-ulose (3-deoxy-D-glucosone), (I), was prepared from ¹³C isotopomers of 3-deoxy-D-ribo-hexose (3-deoxy-D-glucose), (II), using pyranose 2-oxidase (Freimund et al., 1998). Compound (II) crystallizes from water in the β-pyranose form, (III) (Fig. 1), which is the predominant tautomeric form in aqueous solution (~54%) (Pfeffer et al., 1980).

Analysis of the Cremer–Pople puckering parameters (Cremer & Pople, 1975) for (III) and for the related aldohexoses β-D-glucopyranose, (IV) (Kouwijzer et al., 1995), and 2-deoxy-β-D-arabino-hexose, (V) (Maluszynska et al., 1981) (Table 1), shows that all three structures are slightly distorted ⁴C₁ chair forms (q₃ >> q₂). The degree of distortion varies with structure, with θ_(IV) > θ_(III) > θ_(V). The direction of distortion, embodied in the ϕ value, is similar for (IV) and (V), and both differ from (III). The ϕ values suggest a small tendency toward boat-like distortions for (III) and (V), and a twist-boat distortion for (IV) (Fig. 2), based on idealized ϕ values of 60° for (III), 330° for (IV) and 0° for (V).

The structural parameters for compounds (III)–(V) are compared in Table 2. The C—C and C—O bond lengths vary considerably between the structures, with only the C5—C6 distance relatively unchanged. Of the three endocyclic C—C—C bond angles, the C1—C2—C3 angle changes the most (4.1°), and to approximately the same extent as does the exocyclic C4—C5—C6 angle (3.9°). By comparison, the endocyclic C5—O5—C1 angle remains relatively unchanged between the three structures.

The endocyclic torsion angles (absolute values) range from 49.7 to 66.5°, indicative of the non-ideal character of the chair conformations. The C2—C1—O1—H bond torsions are similar in (IV) and (V) (~153°) and are consistent with expectations based on the exoanomeric effect (Lemieux, 1971; Juaristi & Cuevas, 1995). Both are considerably larger than the corresponding torsion in (III) (~111°). The latter torsion indicates a nearly eclipsed conformation for the H1—C1—O1—H torsion angle (−10.9°). Presumably, destabilization caused by eclipsing bonds is overcome by crystal packing forces, to yield this otherwise higher-energy geometry.

The exocyclic hydroxymethyl conformation differs in compounds (III)–(V), with compound (III) favoring the gt conformation (C4 anti to O6), and compounds (IV) and (V) favoring the gg (H5 anti to O6) conformation. In solution, the gg and gt rotamers are considered more favored than the tg rotamer (O5 anti to O6) in aldohexopyranosyl rings containing an equatorial C4—O4 bond (Thibaudeau et al., 2004).

The presence of unquantifiable crystal packing forces complicates the interpretation of structural differences in compounds (III)–(V). Nevertheless, it is instructive to compare compounds (III)–(V) by considering compound (IV) as the parent structure undergoing monodeoxygenation to give (V) or (III). The endocyclic bond lengths C3—C4, C5—C6 and C5—O5 are largely unchanged, and the C1—C2, C2—C3 and C4—C5 bond lengths increase, upon C3-deoxygenation. Only the C1—O5 bond length decreases. In contrast, bond lengths C5—C6, C5—O5 and C1—O5 remain largely unchanged, and C1—C2, C4—C5 and possibly C2—C3 increase, upon C2-deoxygenation. When endocyclic bond-length changes are observed upon monodeoxygenation at equatorial C—O bonds, they more often involve lengthening, i.e. they result in ring expansion/relaxation. The difference between the C1—O5 and C1—O1 bond lengths in (III) is considerably smaller (0.015 Å) than those observed in (IV) and (V) (0.037 and 0.044 Å, respectively), indicating different stereoelectronic properties at the anomeric sites.

The hydrogen bonding in compound (III) (Fig. 3) forms a dense three-dimensional network with no appreciable void space. The network is a 3⁶-net made from four of eight possible connections (four donors and four acceptors). Two of the four connections, O4 and O6, are double bridges, acting as both donor and acceptor. Atom O6, acting as a donor, and atom O2, acting as an acceptor, complete the connections to form the two-dimensional corrugated sheet. The remaining connections, atom O6 as acceptor and atom O2 as donor, link the sheets into a stack. The hydrogen-bonding geometry is summarized in Table 3.

Different H-bonding patterns and/or exocyclic C–O torsions in compounds (III)–(V) may partly explain the bond-length changes observed upon monodeoxygenation. It is noteworthy that, apart from the different hydrogen-bonding properties of atoms O1 and OH4 in compounds (IV) and (V), the hydrogen-bonding and C–O torsional behaviors are otherwise comparable, suggesting that the observed differences in bond lengths are caused mainly by deoxygenation effects. J_CC values in aqueous solution show ¹J_C1,C2 to be larger in (IV) (46.0 Hz) (King-Morris & Serianni, 1987) than in (V) (40.3 Hz) (Bose et al., 1998). A smaller ¹J_CC value would result from the longer C1—C2 bond in (V), superimposed on tdifferences in substituent electronegativity effects at C2.

The differences in hydrogen-bonding patterns and C—O torsion angles are more pervasive in (III) and (IV) and could play a dominant role in determining some of the bond-length differences between these compounds. The latter possibility is supported by ¹J_C1,C2 in (III) (46.9 Hz), which is 0.9 Hz larger than ¹J_C1,C2 in (IV), despite the apparently larger C1—C2 distance observed in the crystal structure. Interestingly, ¹J_C2,C3 is smaller in (III) (35.2 Hz) than in (IV) (38.8 Hz), consistent with the smaller C2—C3 distance in the latter. The significantly different C1—O1 torsion angles in (III) and (IV) (Table 2), and the altered hydrogen-bonding character of atoms O1 and O5, may be responsible for differences in their crystal structures near the anomeric C atom and elsewhere in the structure. The effects of C3-deoxygenation on the anomeric center are probably not long-range intrinsic effects induced solely by monodeoxygenation, but rather are mediated by extrinsic structural changes induced by crystal packing. Presumably, the C1—O1 torsion angles and other extrinsic characteristics in compounds (III) and (IV) are similar in aqueous solution, thus accounting for the comparable ¹J_C1,C2 values observed in the solution state.

S2. Experimental

1,2;5,6-Di-O-isopropylidene-α-D-glucofuranose (5.2 g, 20.0 mmol) was dissolved in tetrahydrofuran (100 ml), imidazole (30 mg) was added, and then NaH (1.0 g, 40 mmol) was added batchwise. The reaction mixture was stirred for 1 h at room temperature under nitrogen. Carbon disulfide (6.0 ml, 100 mmol) was added and the mixture was stirred for 2 h. Methyl iodide (3.0 ml, 48 mmol) was then added, and the mixture was stirred for an additional 1 h. The organic layer was washed with 1 M HCl, saturated NaHCO₃ and brine, and dried over anhydrous Na₂SO₄. After evaporation of the solvent, the residue was recrystallized (EtOH–H₂O Ratio?) to afford 1,2;5,6-di-O-isopropylidene-3-O-(methylthio)thiocarbonyl-α-D-glucofuranose, (VI) (5.59 g, 80%).

Compound (VI) (3.5 g, 10 mmol) in toluene (60 ml) was added by dropping funnel (1 drop every 2 s) to tri-n-butyltin hydride (4.0 g) in toluene (50 ml) under nitrogen and at reflux. Refluxing was continued overnight and the solvent was removed with a rotary evaporator. Acetonitrile (40 ml) and hexane (40 ml) were added to the residue and the two-phase solution was stirred vigorously for 15 min. The lower acetonitrile layer was separated and the hexane phase washed with acetonitrile (15 ml). Extraction of the combined acetonitrile solutions was repeated twice. The combined acetonitrile phase was concentrated to give 1,2;5,6-di-O-isopropylidene-3-deoxy-α-D-glucofuranose, (VII) (2.05 g, 84%).

Compound (VII) (2.0 g, 8.2 mmol) was dissolved in 0.1% (v/v) aqueous H₂SO₄ (120 ml) and the reaction mixture was refluxed for 1 h in an oil bath. After cooling, the solution was treated with Dowex 1 × 8 (HCO₃⁻) ion-exchange resin to adjust the pH to ~7, the resin was removed by filtration, and the solution was concentrated at 303 K in vacuo to give 3-deoxy-D-ribo-hexose, (II) (1.14 g, 85%).

Compound (II) was dissolved in a minimal volume of distilled water and the solution was left at room temperature. Crystals of the β-pyranose, (III), formed slowly and were harvested for structure determination.

S3. Refinement

H atoms were positioned geometrically and treated as riding, with O—H = 0.84 Å and C—H = 0.99–1.00 Å, and with U_iso(H) = 1.2U_eq(C) or 1.5U_eq(O). [Please check added text and correct as necessary]

Computing details

Data collection: APEX2 (Bruker, 2006); cell refinement: APEX2 and SAINT (Bruker, 2006); data reduction: SAINT and XPREP (Sheldrick, 2005); program(s) used to solve structure: XL (Sheldrick, 2001); program(s) used to refine structure: XS (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 molecular structure of (III), with the atom-numbering scheme. Displacement ellipsoids are drawn at the 50% probability level and H atoms are shown as small spheres of arbitrary radii.
	Fig. 2. Ring distortions observed in compounds (III)–(V) based on Cremer–Pople parameters. B denotes the boat form and TB denotes the twist-boat form. The definition of ϕ is given in Cremer & Pople (1975).
	Fig. 3. A packing diagram for (III), viewed down the c axis and parallel to the corrugated sheet structures formed by hydrogen bonding. Dashed lines indicate hydrogen bonding. H atoms have been omitted for clarity.

3-deoxy-β-D-ribo-hexopyranose

Crystal data

C6H12O5	F(000) = 352
Mr = 164.16	Dx = 1.508 Mg m^−3
Orthorhombic, P212121	Cu Kα radiation, λ = 1.54178 Å
Hall symbol: P 2ac 2ab	Cell parameters from 6255 reflections
a = 7.4534 (7) Å	θ = 4.9–69.0°
b = 9.0663 (8) Å	µ = 1.14 mm^−1
c = 10.6966 (9) Å	T = 100 K
V = 722.82 (11) Å^3	Needle, clear colourless
Z = 4	0.21 × 0.19 × 0.19 mm

Data collection

Bruker SMART APEX CCD area-detector diffractometer	1325 independent reflections
Radiation source: fine-focus sealed tube, Siemens KFFCU2K-90	1311 reflections with I > 2σ(I)
Graphite monochromator	Rint = 0.022
Detector resolution: 8.33 pixels mm^-1	θmax = 69.1°, θmin = 6.4°
ω and ϕ scans	h = −8→8
Absorption correction: multi-scan (SADABS; Sheldrick, 2004)	k = 0→10
Tmin = 0.795, Tmax = 0.812	l = 0→12
10176 measured 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)] = 0.026	H-atom parameters constrained
wR(F^2) = 0.069	w = 1/[σ^2(Fo^2) + (0.0403P)^2 + 0.2256P] where P = (Fo^2 + 2Fc^2)/3
S = 1.06	(Δ/σ)max < 0.001
1325 reflections	Δρmax = 0.20 e Å^−3
104 parameters	Δρmin = −0.20 e Å^−3
0 restraints	Absolute structure: Flack (1983), with how many Friedel pairs?
Primary atom site location: structure-invariant direct methods	Absolute structure parameter: 0.03 (19)

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. 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 > 2σ(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
O1	0.63868 (12)	0.87513 (11)	0.74343 (9)	0.0166 (2)
H1O	0.7248	0.8295	0.7762	0.025*
O2	0.47577 (13)	0.94455 (11)	0.98188 (8)	0.0163 (2)
H2O	0.5690	0.9933	0.9664	0.024*
O4	−0.06991 (12)	0.73601 (11)	0.81886 (9)	0.0173 (2)
H4O	−0.0509	0.6827	0.8817	0.026*
O5	0.37441 (13)	0.78829 (11)	0.67501 (8)	0.0162 (2)
O6	0.19681 (13)	0.57721 (11)	0.50954 (9)	0.0178 (2)
H6O	0.1760	0.5864	0.4328	0.027*
C1	0.47656 (18)	0.81383 (15)	0.78443 (12)	0.0144 (3)
H1	0.4991	0.7189	0.8294	0.017*
C2	0.37792 (18)	0.92197 (15)	0.86930 (12)	0.0139 (3)
H2	0.3607	1.0181	0.8251	0.017*
C3	0.19664 (18)	0.85676 (16)	0.90432 (12)	0.0149 (3)
H3B	0.1275	0.9297	0.9535	0.018*
H3Q	0.2148	0.7683	0.9570	0.018*
C4	0.09067 (17)	0.81441 (15)	0.78725 (12)	0.0137 (3)
H4	0.0569	0.9066	0.7416	0.016*
C5	0.20523 (18)	0.71826 (14)	0.70051 (13)	0.0143 (3)
H5	0.2268	0.6204	0.7409	0.017*
C6	0.11368 (18)	0.69546 (15)	0.57620 (12)	0.0157 (3)
H6A	−0.0147	0.6725	0.5899	0.019*
H6B	0.1212	0.7871	0.5262	0.019*

Atomic displacement parameters (Ų)

	U^11	U^22	U^33	U^12	U^13	U^23
O1	0.0109 (4)	0.0210 (5)	0.0180 (5)	−0.0015 (4)	0.0006 (4)	0.0017 (4)
O2	0.0158 (5)	0.0191 (5)	0.0141 (5)	−0.0038 (4)	−0.0007 (4)	0.0003 (4)
O4	0.0114 (5)	0.0224 (5)	0.0182 (5)	−0.0021 (4)	0.0001 (4)	0.0047 (4)
O5	0.0123 (4)	0.0221 (5)	0.0141 (4)	−0.0019 (4)	0.0006 (4)	−0.0015 (4)
O6	0.0196 (5)	0.0203 (5)	0.0136 (4)	0.0041 (4)	−0.0029 (4)	−0.0015 (4)
C1	0.0114 (6)	0.0163 (6)	0.0156 (6)	−0.0017 (5)	−0.0003 (5)	0.0005 (5)
C2	0.0137 (6)	0.0153 (6)	0.0129 (6)	−0.0007 (6)	−0.0010 (5)	0.0008 (5)
C3	0.0133 (6)	0.0167 (6)	0.0146 (6)	0.0019 (5)	0.0018 (5)	0.0001 (5)
C4	0.0101 (6)	0.0160 (6)	0.0150 (6)	−0.0006 (5)	0.0005 (5)	0.0023 (5)
C5	0.0098 (6)	0.0149 (6)	0.0181 (7)	−0.0011 (5)	−0.0012 (5)	0.0011 (5)
C6	0.0145 (6)	0.0174 (7)	0.0153 (6)	0.0011 (6)	−0.0013 (5)	−0.0007 (5)

Geometric parameters (Å, º)

O1—C1	1.4005 (17)	C2—C3	1.5216 (19)
O1—H1O	0.8400	C2—H2	1.0000
O2—C2	1.4227 (15)	C3—C4	1.5295 (18)
O2—H2O	0.8400	C3—H3B	0.9900
O4—C4	1.4325 (16)	C3—H3Q	0.9900
O4—H4O	0.8400	C4—C5	1.5329 (18)
O5—C1	1.4153 (16)	C4—H4	1.0000
O5—C5	1.4379 (16)	C5—C6	1.5088 (18)
O6—C6	1.4289 (16)	C5—H5	1.0000
O6—H6O	0.8400	C6—H6A	0.9900
C1—C2	1.5250 (18)	C6—H6B	0.9900
C1—H1	1.0000
C1—O1—H1O	109.5	C4—C3—H3Q	109.5
C2—O2—H2O	109.5	H3B—C3—H3Q	108.1
C4—O4—H4O	109.5	O4—C4—C3	111.27 (11)
C1—O5—C5	112.78 (10)	O4—C4—C5	109.03 (10)
C6—O6—H6O	109.5	C3—C4—C5	110.53 (11)
O1—C1—O5	105.67 (10)	O4—C4—H4	108.7
O1—C1—C2	110.32 (10)	C3—C4—H4	108.7
O5—C1—C2	109.76 (11)	C5—C4—H4	108.7
O1—C1—H1	110.3	O5—C5—C6	106.86 (11)
O5—C1—H1	110.3	O5—C5—C4	110.62 (10)
C2—C1—H1	110.3	C6—C5—C4	111.06 (11)
O2—C2—C3	107.62 (10)	O5—C5—H5	109.4
O2—C2—C1	110.44 (11)	C6—C5—H5	109.4
C3—C2—C1	108.96 (10)	C4—C5—H5	109.4
O2—C2—H2	109.9	O6—C6—C5	110.27 (11)
C3—C2—H2	109.9	O6—C6—H6A	109.6
C1—C2—H2	109.9	C5—C6—H6A	109.6
C2—C3—C4	110.76 (11)	O6—C6—H6B	109.6
C2—C3—H3B	109.5	C5—C6—H6B	109.6
C4—C3—H3B	109.5	H6A—C6—H6B	108.1
C2—C3—H3Q	109.5
C5—O5—C1—O1	177.22 (10)	C2—C3—C4—C5	51.35 (14)
C5—O5—C1—C2	−63.83 (13)	C1—O5—C5—C6	−178.89 (10)
O1—C1—C2—O2	−66.13 (13)	C1—O5—C5—C4	60.09 (13)
O5—C1—C2—O2	177.83 (10)	O4—C4—C5—O5	−175.25 (9)
O1—C1—C2—C3	175.86 (10)	C3—C4—C5—O5	−52.64 (14)
O5—C1—C2—C3	59.82 (13)	O4—C4—C5—C6	66.27 (13)
O2—C2—C3—C4	−174.16 (11)	C3—C4—C5—C6	−171.12 (11)
C1—C2—C3—C4	−54.39 (14)	O5—C5—C6—O6	74.22 (13)
C2—C3—C4—O4	172.64 (10)	C4—C5—C6—O6	−165.05 (10)

Hydrogen-bond geometry (Å, º)

D—H···A	D—H	H···A	D···A	D—H···A
O1—H1O···O4^i	0.84	1.81	2.6381 (13)	169
O2—H2O···O6^ii	0.84	1.92	2.7222 (14)	159
O4—H4O···O2^iii	0.84	1.87	2.7090 (14)	176
O6—H6O···O1^iv	0.84	1.94	2.7742 (14)	175

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