organic compounds The regioisomeric 1H(2H)-pyrazolo[3,4-d]pyrimidine N 2 -(2 0 -deoxy-b-d-ribofuranosides) Comment

Transcription

1 organic compounds Acta Crystallographica Section C Crystal Structure Communications ISSN stereo-electronic effects of the nucleobases are thought to be responsible for this change. The regioisomeric 1H(2H)-pyrazolo[3,4-d]pyrimidine N 1 - and N 2 -(2 0 -deoxy-b-d-ribofuranosides) Junlin He, a Frank Seela, a * Henning Eickmeier b and Hans Reuter b a Laboratorium fuèr Organische und Bioorganische Chemie, Institut fuèr Chemie, UniversitaÈt OsnabruÈck, Barbarastraûe 7, OsnabruÈck, Germany, and b Anorganische Chemie II, Institut fuèr Chemie, UniversitaÈt OsnabruÈck, Barbarastraûe 7, OsnabruÈck, Germany Correspondence frank.seela@uni-osnabrueck.de Received 28 June 2002 Accepted 19 August 2002 Online 21 September 2002 In the title regioisomeric nucleosides, alternatively called 1-(2- deoxy--d-erythro-furanosyl)-1h-pyrazolo[3,4-d]pyrimidine, C 10 H 12 N 4 O 3, (II), and 2-(2-deoxy--d-erythro-furanosyl)-2Hpyrazolo[3,4-d]pyrimidine, C 10 H 12 N 4 O 3, (III), the conformations of the glycosylic bonds are anti [ (2) for (II) and 15.0 (2) for (III)]. Both nucleosides adopt an S-type sugar pucker, which is C2 0 -endo-c3 0 -exo ( 2 T 3 ) for (II) and 3 0 -exo (between 3 E and 4 T 3 ) for (III). Comment During a search for more stable `da-dt' base pairs, various 3-substituted pyrazolo[3,4-d]pyrimidine 2 0 -deoxyribonucleosides (7-substituted 8-aza-7-deazapurine 2 0 -deoxyribonucleosides) were studied as analogues of 2 0 -deoxyadenosine and were incorporated in oligonucleotides (systematic numbering is used throughout the paper). The interchange of the vemembered ring atoms and the presence of substituents (Br or I) on the 3-position of the modi ed purine bases exert an in uence on the base-pair stability (Seela, Becher & Zulauf, 1999; He & Seela, 2002a). Common 2 0 -deoxyribonucleosides tend to adopt an anti conformation. The orientation of the base relative to the sugar (syn/anti) is de ned by the torsion angle (O4 0 ÐC1 0 ÐN9ÐC4) (purine numbering; IUPAC± IUB Joint Commision on Biochemical Nomenclature, 1983) Deoxyadenosine shows an anti conformation, with a torsion angle (O4 0 ÐC1 0 ÐN1ÐC7a) of (Sato, 1984), while that of pyrazolo[3,4-d]pyrimidin-4-amine 2 0 -deoxyribonucleoside (8-aza-7-deaza-2 0 -deoxyadenosine), (I), is between an anti and a high-anti conformation [ = (2) ; Seela, Zulauf et al., 1999]. Further substitution (Br or I) at the 3-position drives the conformation to high-anti [for the 3-bromo derivative, = 74.1 (4), while for the 3-iodo derivative, = 73.2 (4) ; Seela et al., 2000]. The steric and To the best of our knowledge, there is no reported crystal structure of a pyrazolo[3,4-d]pyrimidin-2-yl 2 0 -deoxyribonucleoside. Here, the X-ray crystallographic analyses of a pair of N 1 - and N 2 -glycosylated pyrazolo[3,4-d]pyrimidines, viz. (II) and (III), respectively, are described. Both nucleosides have the same -d con guration. According to the systematic numbering for compound (II), the torsion angle is de ned by O4 0 ÐC1 0 ÐN1ÐC7a. The de nition of an anti base orientation about the glycosylic bond of the N 2 -nucleoside, (III), is arbitrarily ascribed to the torsion angle O4 0 ÐC1 0 ÐN2ÐC3 of 180, according to Seela & Debelak (2000). From the crystal structure of compound (II) (Fig. 1), the conformation of the glycosylic bond is between the anti and high-anti values [ = (2) ], and is very close to that of compound (I) (Seela, Zulauf et al., 1999). Compound (III) adopts an anti conformation, with = 15.0 (2). The glycosylic bond between atoms N2 and C1 0 of compound (III) is A Ê longer than that between atoms N1 and C1 0 of compound (II). Both nucleosides show an S-type sugar conformation, but with different ring puckering. The sugar conformation of nucleoisde (II) is C2 0 -endo-c3 0 -exo ( 2 T 3 ), with pseudo-rotation parameters (Rao et al., 1981) P = (2) and m = 40.3 (1), while the sugar part of nucleoside (III) has a 3 0 -exo conformation (between 3 E and 4 T 3 ), with pseudo-rotation parameters P = (1) and m = 37.0 (1). These two nucleosides have the same ap (g ) conformation about the C4 0 ÐC5 0 bond; the values of (C3 0 ÐC4 0 ÐC5 0 ÐO5 0 ) are (18) and (13) for (II) and (III), respectively. This means that the base and the hydroxymethyl group undergo the same disrotatory motion so that the Coulombic repulsion between atoms N2 and O5 0 or between atoms N1 and O4 0 is minimized (Seela, Becher et al., 1999). Similarly, nucleoside (I) adopts the C2 0 -endo-c3 0 -exo-type (S-type) sugar puckering, but with a ap conformation around the C4 0 ÐC5 0 bond [ = (16) ; Seela, Zulauf et al., 1999], while 2 0 -deoxyadenosine has a C3 0 -endo conformation (Sato, 1984). These results have an in uence on the stability of oligonucleotide duplexes (He & Seela, 2002a). The base moieties of (II) and (III) are nearly planar. The r.m.s. deviations of the ring atoms (N1/N2/C3/C3a/C4/N5/C6/ N7/C7a) from their calculated least-squares planes are and A Ê, respectively, with the maximum deviations being (2) (N1) and (1) A Ê (C3a). Atom C1 0 is displaced from this plane by (2) and (1) A Ê in (II) and (III), Acta Cryst. (2002). C58, o593±o595 DOI: /S # 2002 International Union of Crystallography o593

2 organic compounds m.p.: 421 K; R F (silica-gel thin-layer chromatography): 0.22 (CH 2 Cl 2 / CH 3 OH, 9:1). Suitable crystals were grown from a solution in methanol. Nucleoside (III) was obtained as the minor product from the above glycosylation reaction followed by deprotection of the sugar moiety; m.p.: 427 K; R F (silica-gel thin-layer chromatography): 0.13 (CH 2 Cl 2 /CH 3 OH, 9:1). Suitable crystals were grown from a solution in acetone. Compound (II) Crystal data C 10 H 12 N 4 O 3 M r = Orthorhombic, P a = (8) A Ê b = (15) A Ê c = (11) A Ê V = (2) A Ê 3 Z =4 D x = Mg m 3 Data collection Bruker P4 diffractometer 2/! scans 2375 measured re ections 1784 independent re ections 1474 re ections with I > 2(I) R int = max = 30.0 Mo K radiation Cell parameters from 40 re ections = 6.9±12.5 = 0.11 mm 1 T = 293 (2) K Transparent needle, colourless mm h = 1! 9 k = 15! 1 l = 19! 1 3 standard re ections every 97 re ections intensity decay: none Figure 1 Perspective views of nucleosides (a) (II) and (b) (III). Displacement ellipsoids for non-h atoms are drawn at the 50% probability level and H atoms are shown as small spheres of arbitrary size. respectively. The bases are strongly stacked in both crystal structures. Structures (II) and (III), which differ only in the glycosylation positions (N1 versus N2), each form two different types of hydrogen bonds. Structure (II) is stabilized by intermolecular hydrogen bonds between O3 0 ÐH and O4 0 (2 x, y 1 2, 3 2 z) of two sugar moieties and between O50 ÐOH of the sugar moiety and N7(1 + x, y, z) of an adjacent nucleobase unit. These interactions link the molecules into an in nite twodimensional network in which the bases are stacked and tilted with respect to each other. In contrast, structure (III) is stabilized exclusively by hydrogen bonds between the bases and sugar units [O3 0 1 ÐH with N7(1 x, 2 + y, 2 z) and O5 0 ÐH with N5(2 x, y, 1 z)]. These interactions link the molecules into an in nite two-dimensional network, with piles of stacked bases tilted only slightly with respect to each other. Experimental Nucleoside (II) (Seela & Steker, 1984) was prepared by the glycosylation reaction of pyrazolo[3,4-d]pyrimidine with 2-deoxy-3,5-di- O-(p-toluoyl)--d-erythro-furanosyl chloride (Hoffer, 1960), followed by deprotection of the sugar moiety (He & Seela, 2002b); Re nement Re nement on F 2 R[F 2 >2(F 2 )] = wr(f 2 ) = S = re ections 164 parameters H atoms treated by a mixture of independent and constrained re nement Table 1 Selected geometric parameters (A Ê, ) for (II). N1ÐC (2) w = 1/[ 2 (F o 2 ) + (0.0518P) P] where P =(F o 2 +2F c 2 )/3 (/) max < max = 0.21 e A Ê 3 min = 0.18 e A Ê 3 Extinction correction: SHELXTL Extinction coef cient: (3) C7aÐN1ÐC (18) N2ÐN1ÐC (17) C7aÐN1ÐN2ÐC3 0.8 (3) C1 0 ÐN1ÐN2ÐC (2) C4ÐC3aÐC7aÐN7 1.4 (3) C3ÐC3aÐC7aÐN (19) C7aÐN1ÐC1 0 ÐO (2) N2ÐN1ÐC1 0 ÐO (2) C7aÐN1ÐC1 0 ÐC (2) O4 0 ÐC1 0 ÐC2 0 ÐC (2) N1ÐC1 0 ÐC2 0 ÐC (19) Table 2 Hydrogen-bonding geometry (A Ê, ) for (II). C1 0 ÐC2 0 ÐC3 0 ÐO (2) C1 0 ÐC2 0 ÐC3 0 ÐC (2) C2 0 ÐC3 0 ÐC4 0 ÐO (2) O3 0 ÐC3 0 ÐC4 0 ÐC (17) N1ÐC1 0 ÐO4 0 ÐC (17) C5 0 ÐC4 0 ÐO4 0 ÐC (19) C3 0 ÐC4 0 ÐO4 0 ÐC (2) O4 0 ÐC4 0 ÐC5 0 ÐO (2) C3 0 ÐC4 0 ÐC5 0 ÐO (18) DÐHA DÐH HA DA DÐHA O3 0 ÐH3 0 OO4 0i 0.80 (2) 2.16 (2) (2) 149 (3) O5 0 ÐH5 0 ON7 ii 0.80 (2) 2.04 (2) (3) 168 (3) Symmetry codes: (i) 2 x; y 1 2 ; 3 2 z; (ii) 1 x; y; z. o594 Junlin He et al. Two regioisomers of C 10 H 12 N 4 O 3 Acta Cryst. (2002). C58, o593±o595

3 organic compounds Table 3 Selected geometric parameters (A Ê, ) for (III). N2ÐC (18) C7aÐN1ÐN (12) C3ÐN2ÐC (13) C7aÐN1ÐN2ÐC (17) C7aÐN1ÐN2ÐC (12) C3ÐC3aÐC7aÐN (14) C3ÐN2ÐC1 0 ÐO (2) N1ÐN2ÐC1 0 ÐO (12) N1ÐN2ÐC1 0 ÐC (16) O4 0 ÐC1 0 ÐC2 0 ÐC (15) N2ÐC1 0 ÐC2 0 ÐC (12) Compound (III) Crystal data C 10 H 12 N 4 O 3 M r = Monoclinic, P2 1 a = (7) A Ê b = (14) A Ê c = (12) A Ê = (9) V = (12) A Ê 3 Z =2 Data collection Bruker P4 diffractometer 2/! scans 2202 measured re ections 1562 independent re ections 1524 re ections with I > 2(I) R int = max = 30.0 Re nement Re nement on F 2 R[F 2 >2(F 2 )] = wr(f 2 ) = S = re ections 164 parameters H atoms treated by a mixture of independent and constrained re nement Table 4 Hydrogen-bonding geometry (A Ê, ) for (III). N1ÐN2ÐC (12) C1 0 ÐC2 0 ÐC3 0 ÐO (15) C1 0 ÐC2 0 ÐC3 0 ÐC (15) C2 0 ÐC3 0 ÐC4 0 ÐO (14) O3 0 ÐC3 0 ÐC4 0 ÐC (12) C2 0 ÐC1 0 ÐO4 0 ÐC (14) O4 0 ÐC4 0 ÐC5 0 ÐO (17) C3 0 ÐC4 0 ÐC5 0 ÐO (13) D x = Mg m 3 Mo K radiation Cell parameters from 47 re ections = 4.5±16.1 = 0.12 mm 1 T = 293 (2) K Transparent block, yellow mm h = 6! 1 k = 1! 18 l = 11! 11 3 standard re ections every 97 re ections intensity decay: none w = 1/[ 2 (F o 2 ) + (0.0637P) P] where P =(F o 2 +2F c 2 )/3 (/) max < max = 0.28 e A Ê 3 min = 0.20 e A Ê 3 Extinction correction: SHELXTL Extinction coef cient: (9) DÐHA DÐH HA DA DÐHA O3 0 ÐH3 0 ON7 i (15) (15) (2) 175 (2) O5 0 ÐH5 0 ON5 ii (15) (15) (19) (17) Symmetry codes: (i) 1 x; 1 2 y; 2 z; (ii) 2 x; 1 2 y; 1 z. In the absence of suitable anomalous scattering, Friedel equivalents could not be used to determine the absolute structure. Re nement of the Flack (1983) parameter led to inconclusive values (Flack & Bernadinelli, 2000) [ 0.2 (16) for (II) and 0.4 (10) for (III)]. Therefore, the Friedel equivalents [416 for (II) and 119 for (III)] were merged before the nal re nements. The known con guration of the parent molecule was used to de ne the enantiomer employed in the re ned model. All H atoms were initially found in a difference Fourier synthesis. In order to maximize the data-to-parameter ratio, the H atoms bonded to C atoms were placed in geometrically idealized positions (CÐH = 0.93±0.98 A Ê ) and constrained to ride on their parent atoms. The hydroxy H atoms were initially placed in difference-map positions, then geometrically idealized and constrained to ride on their parent O atoms, although the chemically equivalent OÐ H bond lengths were allowed to re ne while being constrained to be equal. An overall isotropic displacement parameter was re ned for all H atoms. For both compounds, data collection: XSCANS (Siemens, 1996); cell re nement: XSCANS; data reduction: SHELXTL (Sheldrick, 1997); program(s) used to solve structure: SHELXTL; program(s) used to re ne structure: SHELXTL; molecular graphics: SHELXTL; software used to prepare material for publication: SHELXTL. Supplementary data for this paper are available from the IUCr electronic archives (Reference: LN1143). Services for accessing these data are described at the back of the journal. References Flack, H. D. (1983). Acta Cryst. A39, 876±881. Flack, H. D. & Bernadinelli, G. (2000). J. Appl. Cryst. 33, 1143±1148. He, J. & Seela, F. (2002a). Tetrahedron, 58, 4535±4542. He, J. & Seela, F. (2002b). In preparation. Hoffer, M. (1960). Chem. Ber. 93, 2777±2781. IUPAC±IUB Joint Commission on Biochemical Nomenclature (1983). Eur. J. Biochem. 131, 9±15. Rao, S. T., Westhof, E. & Sundaralingam, M. (1981). Acta Cryst. A37, 421± 425. Sato, T. (1984). Acta Cryst. C40, 880±882. Seela, F., Becher, G., Rosemeyer, H., Reuter, H., Kastner, G. & Mikhailopulo, I. A. (1999). Helv. Chim. Acta, 82, 105±124. Seela, F., Becher, G. & Zulauf, M. (1999). Nucleosides Nucleotides, 18, 1399± Seela, F. & Debelak, H. (2000). Nucleic Acids Res. 28, 3224±3232. Seela, F. & Steker, H. (1984). Liebigs Ann. Chem. pp. 1719±1730. Seela, F., Zulauf, M., Reuter, H. & Kastner, G. (1999). Acta Cryst. C55, 1947± Seela, F., Zulauf, M., Reuter, H. & Kastner, G. (2000). Acta Cryst. C56, 489± 491. Sheldrick, G. M. (1997). SHELXTL. Release 5.1. Bruker AXS Inc., Madison, Wisconsin, USA. Siemens (1996). XSCANS. Release 2.2. Siemens Analytical X-ray Instruments Inc., Madison, Wisconsin, USA. Acta Cryst. (2002). C58, o593±o595 Junlin He et al. Two regioisomers of C 10 H 12 N 4 O 3 o595

4 supporting information [doi: /s ] The regioisomeric 1H(2H)-pyrazolo[3,4-d]pyrimidine N 1 - and N 2 -(2 -deoxy-β-dribofuranosides) Junlin He, Frank Seela, Henning Eickmeier and Hans Reuter Computing details For both compounds, data collection: XSCANS (Siemens, 1996); cell refinement: XSCANS; data reduction: SHELXTL (Sheldrick, 1997); program(s) used to solve structure: SHELXTL; program(s) used to refine structure: SHELXTL; molecular graphics: SHELXTL; software used to prepare material for publication: SHELXTL. (II) 1-(2-deoxy-β-D-erythro-furanosyl)-1H-pyrazolo[3,4-d]pyrimidine Crystal data C 10 H 12 N 4 O 3 M r = Orthorhombic, P Hall symbol: P 2ac 2ab a = (8) Å b = (15) Å c = (11) Å V = (2) Å 3 Z = 4 F(000) = 496 Data collection Bruker P4 diffractometer Radiation source: fine-focus sealed tube Graphite monochromator 2θ/ω scans 2375 measured reflections 1784 independent reflections 1474 reflections with I > 2σ(I) Refinement Refinement on F 2 Least-squares matrix: full R[F 2 > 2σ(F 2 )] = wr(f 2 ) = S = reflections 164 parameters 4 restraints Primary atom site location: structure-invariant direct methods D x = Mg m 3 Melting point: K Mo Kα radiation, λ = Å Cell parameters from 40 reflections θ = µ = 0.11 mm 1 T = 293 K Transparent needle, colourless mm R int = θ max = 30.0, θ min = 2.4 h = 1 9 k = 15 1 l = standard reflections every 97 reflections intensity decay: none Secondary atom site location: difference Fourier map Hydrogen site location: inferred from neighbouring sites H atoms treated by a mixture of independent and constrained refinement w = 1/[σ 2 (F o2 ) + (0.0518P) P] where P = (F o 2 + 2F c2 )/3 (Δ/σ) max < Δρ max = 0.21 e Å 3 sup-1

5 Δρ min = 0.18 e Å 3 Extinction correction: SHELXTL, Fc * =kfc[ xfc 2 λ 3 /sin(2θ)] -1/4 Extinction coefficient: (3) 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 F 2 against ALL reflections. The weighted R-factor wr and goodness of fit S are based on F 2, conventional R-factors R are based on F, with F set to zero for negative F 2. The threshold expression of F 2 > σ(f 2 ) is used only for calculating R-factors(gt) etc. and is not relevant to the choice of reflections for refinement. R-factors based on F 2 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 (Å 2 ) x y z U iso */U eq N (3) (15) (12) (4) N (3) (19) (14) (5) C (4) (2) (16) (6) H3A (2)* C3A (3) (19) (15) (5) C (4) (2) (17) (6) H4A (2)* N (3) (18) (15) (5) C (4) (2) (18) (6) H6A (2)* N (3) (16) (12) (4) C7A (3) (16) (13) (4) C (3) (17) (14) (4) H1 A (2)* C (4) (17) (15) (5) H2 A (2)* H2 B (2)* O (3) (14) (14) (5) H3 O (4) (19) (17) (2)* C (3) (16) (15) (5) H3 C (2)* C (3) (16) (13) (4) H (2)* O (3) (11) (10) (4) C (4) (2) (16) (5) H5 A (2)* H5 B (2)* O (3) (2) (12) (6) H5 O (4) (2) (16) (2)* sup-2

6 Atomic displacement parameters (Å 2 ) U 11 U 22 U 33 U 12 U 13 U 23 N (9) (8) (8) (8) (8) (7) N (10) (11) (9) (10) (9) (9) C (13) (13) (10) (12) (11) (10) C3A (12) (10) (9) (9) (10) (8) C (15) (10) (11) (11) (11) (10) N (13) (10) (11) (10) (11) (9) C (13) (11) (13) (11) (13) (11) N (9) (8) (8) (8) (8) (7) C7A (10) (8) (8) (8) (8) (8) C (10) (8) (9) (8) (9) (8) C (13) (9) (10) (10) (10) (8) O (12) (7) (11) (8) (11) (8) C (11) (7) (10) (8) (10) (8) C (11) (8) (8) (8) (9) (7) O (10) (6) (7) (7) (8) (5) C (12) (12) (10) (11) (11) (10) O (14) (13) (8) (12) (10) (9) Geometric parameters (Å, º) N1 C7A (3) C1 H1 A N1 N (2) C2 C (3) N1 C (2) C2 H2 A N2 C (3) C2 H2 B C3 C3A (3) O3 C (3) C3 H3A O3 H3 O 0.80 (2) C3A C (3) C3 C (3) C3A C7A (3) C3 H3 C C4 N (3) C4 O (2) C4 H4A C4 C (3) N5 C (3) C4 H C6 N (3) C5 O (3) C6 H6A C5 H5 A N7 C7A (3) C5 H5 B C1 O (2) O5 H5 O 0.80 (2) C1 C (3) C7A N1 N (16) C1 C2 C (15) C7A N1 C (18) C1 C2 H2 A N2 N1 C (17) C3 C2 H2 A C3 N2 N (19) C1 C2 H2 B N2 C3 C3A (2) C3 C2 H2 B N2 C3 H3A H2 A C2 H2 B C3A C3 H3A C3 O3 H3 O (16) C4 C3A C7A (2) O3 C3 C (17) sup-3

7 C4 C3A C (2) O3 C3 C (2) C7A C3A C (19) C4 C3 C (15) N5 C4 C3A (2) O3 C3 H3 C N5 C4 H4A C4 C3 H3 C C3A C4 H4A C2 C3 H3 C C4 N5 C (2) O4 C4 C (16) N7 C6 N (2) O4 C4 C (16) N7 C6 H6A C5 C4 C (17) N5 C6 H6A O4 C4 H C6 N7 C7A (19) C5 C4 H N7 C7A N (18) C3 C4 H N7 C7A C3A (19) C1 O4 C (14) N1 C7A C3A (19) O5 C5 C (19) O4 C1 N (16) O5 C5 H5 A O4 C1 C (16) C4 C5 H5 A N1 C1 C (16) O5 C5 H5 B O4 C1 H1 A C4 C5 H5 B N1 C1 H1 A H5 A C5 H5 B C2 C1 H1 A C5 O5 H5 O (16) C7A N1 N2 C3 0.8 (3) C3 C3A C7A N1 1.9 (2) C1 N1 N2 C (2) C7A N1 C1 O (2) N1 N2 C3 C3A 0.5 (3) N2 N1 C1 O (2) N2 C3 C3A C (3) C7A N1 C1 C (2) N2 C3 C3A C7A 1.5 (3) N2 N1 C1 C (3) C7A C3A C4 N5 0.4 (3) O4 C1 C2 C (2) C3 C3A C4 N (3) N1 C1 C2 C (19) C3A C4 N5 C6 1.0 (3) C1 C2 C3 O (2) C4 N5 C6 N7 1.6 (4) C1 C2 C3 C (2) N5 C6 N7 C7A 0.6 (3) O3 C3 C4 O (18) C6 N7 C7A N (2) C2 C3 C4 O (2) C6 N7 C7A C3A 1.0 (3) O3 C3 C4 C (17) N2 N1 C7A N (19) C2 C3 C4 C (2) C1 N1 C7A N7 5.5 (3) N1 C1 O4 C (17) N2 N1 C7A C3A 1.8 (2) C2 C1 O4 C4 9.1 (2) C1 N1 C7A C3A (2) C5 C4 O4 C (19) C4 C3A C7A N7 1.4 (3) C3 C4 O4 C (2) C3 C3A C7A N (19) O4 C4 C5 O (2) C4 C3A C7A N (18) C3 C4 C5 O (18) Hydrogen-bond geometry (Å, º) D H A D H H A D A D H A O3 H3 O O4 i 0.80 (2) 2.16 (2) (2) 149 (3) O5 H5 O N7 ii 0.80 (2) 2.04 (2) (3) 168 (3) Symmetry codes: (i) x+2, y 1/2, z+3/2; (ii) x+1, y, z. sup-4

8 (III) 2-(2-deoxy-β-D-erythro-furanosyl)-2H-pyrazolo[3,4-d]pyrimidine Crystal data C 10 H 12 N 4 O 3 M r = Monoclinic, P2 1 Hall symbol: P 2yb a = (7) Å b = (14) Å c = (12) Å β = (9) V = (12) Å 3 Z = 2 Data collection Bruker P4 diffractometer Radiation source: fine-focus sealed tube Graphite monochromator 2θ/ω scans 2202 measured reflections 1562 independent reflections 1524 reflections with I > 2σ(I) Refinement Refinement on F 2 Least-squares matrix: full R[F 2 > 2σ(F 2 )] = wr(f 2 ) = S = reflections 164 parameters 51 restraints Primary atom site location: structure-invariant direct methods Secondary atom site location: difference Fourier map F(000) = 248 D x = Mg m 3 Melting point: K Mo Kα radiation, λ = Å Cell parameters from 47 reflections θ = µ = 0.12 mm 1 T = 293 K Transparent block, yellow mm R int = θ max = 30.0, θ min = 3.0 h = 6 1 k = 1 18 l = standard reflections every 97 reflections intensity decay: none Hydrogen site location: inferred from neighbouring sites H atoms treated by a mixture of independent and constrained refinement w = 1/[σ 2 (F o2 ) + (0.0637P) P] where P = (F o 2 + 2F c2 )/3 (Δ/σ) max < Δρ max = 0.28 e Å 3 Δρ min = 0.20 e Å 3 Extinction correction: SHELXTL, Fc * =kfc[ xfc 2 λ 3 /sin(2θ)] -1/4 Extinction coefficient: (9) 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 F 2 against ALL reflections. The weighted R-factor wr and goodness of fit S are based on F 2, conventional R-factors R are based on F, with F set to zero for negative F 2. The threshold expression of F 2 > σ(f 2 ) is used only for calculating R-factors(gt) etc. and is not relevant to the choice of reflections for refinement. R-factors based on F 2 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 (Å 2 ) x y z U iso */U eq N (3) (10) (16) (3) N (3) (10) (15) (3) sup-5

9 C (3) (12) (17) (3) H (17)* C3A (3) (12) (17) (3) C (3) (13) (18) (3) H (17)* N (3) (12) (17) (3) C (4) (15) (2) (4) H (17)* N (3) (12) (18) (3) C7A (3) (11) (17) (3) C (3) (11) (17) (3) H1 C (17)* C (3) (13) (18) (3) H2 A (17)* H2 B (17)* C (3) (12) (17) (3) H3 C (17)* O (3) (10) (17) (3) H3 O (4) (14) (3) (17)* C (3) (11) (17) (3) H4O (17)* O (2) (9) (13) (2) C (3) (13) (19) (3) H5 B (17)* H5 C (17)* O (3) (14) (19) (4) H5 O (4) (2) (2) (17)* Atomic displacement parameters (Å 2 ) U 11 U 22 U 33 U 12 U 13 U 23 N (7) (6) (6) (5) (5) (4) N (6) (6) (5) (4) (4) (4) C (6) (6) (6) (5) (5) (5) C3A (6) (6) (5) (5) (5) (5) C (7) (8) (6) (6) (5) (5) N (6) (7) (6) (6) (5) (6) C (9) (8) (7) (7) (7) (6) N (8) (6) (6) (6) (6) (5) C7A (6) (6) (6) (5) (5) (5) C (6) (6) (6) (5) (5) (5) C (7) (6) (6) (6) (5) (5) C (6) (6) (5) (5) (5) (5) O (7) (6) (6) (5) (5) (5) C (5) (6) (6) (4) (5) (5) O (5) (5) (4) (4) (4) (4) C (6) (7) (7) (5) (5) (6) O (6) (11) (8) (7) (5) (7) sup-6

10 Geometric parameters (Å, º) N1 C7A (18) C1 H1 C N1 N (18) C2 C (2) N2 C (18) C2 H2 A N2 C (18) C2 H2 B C3 C3A (2) C3 O (18) C3 H C3 C (2) C3A C (19) C3 H3 C C3A C7A (2) O3 H3 O (15) C4 N (2) C4 O (17) C4 H C4 C (2) N5 C (2) C4 H4O C6 N (2) C5 O (2) C6 H C5 H5 B N7 C7A (2) C5 H5 C C1 O (19) O5 H5 O (15) C1 C (2) C7A N1 N (12) C3 C2 C (12) C3 N2 N (12) C3 C2 H2 A C3 N2 C (13) C1 C2 H2 A N1 N2 C (12) C3 C2 H2 B N2 C3 C3A (13) C1 C2 H2 B N2 C3 H H2 A C2 H2 B C3A C3 H O3 C3 C (13) C3 C3A C (15) O3 C3 C (13) C3 C3A C7A (12) C2 C3 C (11) C4 C3A C7A (14) O3 C3 H3 C N5 C4 C3A (15) C2 C3 H3 C N5 C4 H C4 C3 H3 C C3A C4 H C3 O3 H3 O (12) C4 N5 C (13) O4 C4 C (11) N7 C6 N (16) O4 C4 C (12) N7 C6 H C5 C4 C (12) N5 C6 H O4 C4 H4O C6 N7 C7A (15) C5 C4 H4O N1 C7A N (13) C3 C4 H4O N1 C7A C3A (13) C1 O4 C (10) N7 C7A C3A (13) O5 C5 C (12) O4 C1 N (11) O5 C5 H5 B O4 C1 C (11) C4 C5 H5 B N2 C1 C (12) O5 C5 H5 C O4 C1 H1 C C4 C5 H5 C N2 C1 H1 C H5 B C5 H5 C C2 C1 H1 C C5 O5 H5 O (12) C7A N1 N2 C (17) C3 N2 C1 O (2) sup-7

11 C7A N1 N2 C (12) N1 N2 C1 O (12) N1 N2 C3 C3A 0.39 (17) C3 N2 C1 C (17) C1 N2 C3 C3A (13) N1 N2 C1 C (16) N2 C3 C3A C (17) O4 C1 C2 C (15) N2 C3 C3A C7A 0.44 (16) N2 C1 C2 C (12) C3 C3A C4 N (17) C1 C2 C3 O (15) C7A C3A C4 N5 1.7 (2) C1 C2 C3 C (15) C3A C4 N5 C6 0.0 (2) O3 C3 C4 O (13) C4 N5 C6 N7 2.3 (3) C2 C3 C4 O (14) N5 C6 N7 C7A 2.3 (3) O3 C3 C4 C (12) N2 N1 C7A N (15) C2 C3 C4 C (15) N2 N1 C7A C3A 0.17 (17) N2 C1 O4 C (12) C6 N7 C7A N (16) C2 C1 O4 C (14) C6 N7 C7A C3A 0.2 (2) C5 C4 O4 C (14) C3 C3A C7A N (18) C3 C4 O4 C (14) C4 C3A C7A N (13) O4 C4 C5 O (17) C3 C3A C7A N (14) C3 C4 C5 O (13) C4 C3A C7A N7 1.7 (2) Hydrogen-bond geometry (Å, º) D H A D H H A D A D H A O3 H3 O N7 i 0.84 (2) 1.99 (2) (2) 175 (2) O5 H5 O N5 ii 0.84 (2) 1.97 (2) (19) 168 (2) Symmetry codes: (i) x+1, y+1/2, z+2; (ii) x+2, y+1/2, z+1. sup-8