Received 21 January 2011
aLaboratory of Bioorganic Chemistry and Chemical Biology, Center for Nanotechnology, Heisenbergstrasse 11, 48149 Münster, Germany, and Laboratorium für Organische und Bioorganische Chemie, Institut für Chemie, Universität Osnabrück, Barbarastrasse 7, 49069 Osnabrück, Germany,bLaboratory of Bioorganic Chemistry and Chemical Biology, Center for Nanotechnology, Heisenbergstrasse 11, 48149 Münster, Germany, and cAnorganische Chemie II, Institut für Chemie, Universität Osnabrück, Barbarastrasse 7, 49069 Osnabrück, Germany
The title compound [systematic name: 7-(2-deoxy--D-erythro-pentofuranosyl)-7H-imidazo[1,2-c]pyrrolo[2,3-d]pyrimidine hemihydrate], 2C13H14N4O3·H2O or (I)·0.5H2O, shows two similar conformations in the asymmetric unit. These two conformers are connected through one water molecule by hydrogen bonds. The N-glycosylic bonds of both conformers show an almost identical anti conformation with = -107.7 (2)° for conformer (I-1) and -107.0 (2)° for conformer (I-2). The sugar moiety adopts an unusual N-type (C3'-endo) sugar pucker for 2'-deoxyribonucleosides, with P = 36.8 (2)° and m = 40.6 (1)° for conformer (I-1), and P = 34.5 (2)° and m = 41.4 (1)° for conformer (I-2). Both conformers and the solvent molecule participate in the formation of a three-dimensional pattern with a `chain'-like arrangement of the conformers. The structure is stabilized by intermolecular O-HO and O-HN hydrogen bonds, together with weak C-HO contacts.
Etheno adducts have proved to be biomarkers for DNA damage arising from reactions of endogenous lipid peroxidation, chloroethylene oxide or chloroacetaldehyde (Bolt, 1994). They are also thought to initiate vinyl-chloride- and urethane-induced tumours because of their miscoding capability, leading to point mutations (Arab et al., 2009; Pandya & Moriya, 1996). 1,N6-Etheno-2'-deoxytubercidin, (I), and the corresponding congener 1,N6-ethenoadenosine, (III), can be considered as 7-deazapurine or purine pyrrole ring annelation products with a [1,2-c]-ring connectivity (purine numbering is used throughout this discussion). By enlarging the aromatic system, these tricyclic nucleosides show strong fluorescence with quantum yields higher than 0.5 (Seela et al., 2007). Their propensity to fluorescence makes these compounds valuable for probing the biochemical and biophysical properties of nucleosides, nucleotides and nucleic acids (Bielecki et al., 2000; Inoue et al., 1981; Paulsen & Wintermeyer, 1984; Secrist et al., 1972; Seela et al., 2007). The 7-deazapurine nucleoside, (I), shows extraordinary stability in acidic and in alkaline media compared to its `purine' counterpart, (III) (Seela et al., 2007). The synthesis of the title compound, (I), which was prepared from 2'-deoxytubercidin with chloroacetaldehyde, was reported previously (Seela et al., 2007). The single-crystal structure of (I) is studied herein and is compared to the closely related crystal structures of 2'-deoxytubercidin [(IIa) and (IIb); Zabel et al., 1987], 1,N6-ethenoadenosine [(III); Jaskólski, 1982] and 7-deaza-2,8-diaza-1,N6-ethenoadenosine [(IV); Lin et al., 2004].
In the asymmetric unit of (I)·0.5H2O, two conformers with a slightly different sugar puckering exist which are connected through a water molecule by hydrogen bonds. They are defined as types 1 and 2, and denoted (I-1) and (I-2), respectively. The three-dimensional structures of the molecules of (I-1) and (I-2) are shown in Figs. 1 and 2, and selected geometric parameters are summarized in Table 1.
Conformers (I-1) and (I-2) exhibit almost identical torsion angles (O4'-C1'-N9-C4) of -107.7 (2) and -107.0 (2)°, respectively, which indicate conformations situated between anti and high-anti (IUPAC-IUB Joint Commission on Biochemical Nomenclature, 1983). These values are close to that of the water-free crystal of (IIa) [ = -104.4 (2)°], whereas the torsion angle of dihydrate (IIb) [ = -115.5 (3)°] falls into the anti range (Zabel et al., 1987). The length of the glycosylic N9-C1' bond is 1.451 (2) Å for (I-1) and 1.449 (2) Å for (I-2), which is almost identical to the bond length observed for 2'-deoxytubercidin [1.449 (2) Å in (IIa) and 1.446 (4) Å in (IIb); Zabel et al., 1987]. The parent ribonucleoside, (III), adopts a slightly longer glycosylic bond [1.455 (4) Å; Jaskólski, 1982].
The heterocyclic base moiety of 1,N6-ethenoadenosine, (III), forms a `U'-shaped structure when looking from the edge side, with a maximum deviation of 0.064 (4) Å out of the plane (Jaskólski, 1982). In contrast, the 7-deazapurine moieties of (I-1) and (I-2) are nearly planar. The r.m.s. deviations of the ring atoms from their calculated least-squares planes are 0.0121 Å for (I-1) and 0.0206 Å for (I-2). Maximum deviations of 0.0185 (2) and 0.0365 (2) Å were found for atom C112 of (I-1) and atom N29 of (I-2), respectively.
For both conformers, the torsion angle about the exocyclic C4'-C5' bond, which is defined as (O5'-C5'-C4'-C3'), adopts an antiperiplanar (gauche, trans) conformation with = -168.7 (2)° for (I-1) and = -167.1 (2)° for (I-2). In the crystal structures of (IIa) and (IIb), the torsion angles are also within the antiperiplanar range [-179.6 (2) and -173.6 (3)°; trans] (Zabel et al., 1987).
Usually, the sugar conformation of ribonucleosides adopts the N-type pucker, whereas 2'-deoxyribonucleosides prefer the S conformation. In solution, the predominant conformation of compound (I) shows the S-type conformation (75% S). The sugar conformation of compound (I) was determined from the vicinal 3J(H,H) coupling constants of the 1H NMR spectra measured in D2O, applying the program PSEUROT6.3 (Van Wijk et al., 1999). It has to be noted that both conformers exhibit sugar moieties with the N conformation in the crystalline state. For conformer (I-1), the sugar pucker is 4T3 (C4'-exo-C3'-endo) (Altona & Sundaralingam, 1972), with a phase angle of pseudorotation of P = 36.8 (2)° and a maximum amplitude of puckering of m = 40.6 (1)°. In conformer (I-2), the sugar moiety adopts a slightly different N-type sugar pucker (3T4; C3'-endo-C4'-exo), with P = 34.5 (2)° and m = 41.4 (1)°. In contrast, the parent 2'-deoxytubercidins, (IIa) and (IIb), adopt S conformations with P = 186.6 (2) (3T2; C3'-exo-C2'-endo) and 215.1 (3)° (3T4; C3'-exo-C4'-endo), respectively. A similar influence on the sugar conformation was also found for the ribonucleoside 1,N6-etheno derivatives, (III) and (IV), which adopt the S conformation (C2'-endo) instead of the usual N-type conformation of ribonucleosides. The ribose ring of nucleoside (III) is characterized by P = 163.5° (2T3; C2'-endo-C3'-exo) and m = 44.3° (Jaskólski, 1982), while P = 183.4° (3T2; C3'-exo-C2'-endo) and m = 42.4° for compound (IV) (Lin et al., 2004).
The title compound forms a three-dimensional network, which is generated by numerous hydrogen bonds involving conformers (I-1) and (I-2) and the water molecule (Fig. 3 and Table 2). Within the ac plane, (I-1) and (I-2) are located in a `chain'-like arrangement. Each chain is composed of molecules of identical conformation, either (I-1) or (I-2), and the chains are ordered in an alternating fashion. Furthermore, within the chains, the individual molecules are arranged in a head-to-tail fashion. The different chains are connected to each other via hydrogen bonding between the two conformers. The individual chains are also stabilized by hydrogen bonds, while the water molecule participates in both intra- and interchain hydrogen bonds. Conformers (I-1) and (I-2) show a different hydrogen-bonding pattern. Hydrogen bonds are formed to neighbouring molecules of identical conformation (O13'-H13'O15'i, O15'-H15'N112ii, O23'-H23'O25'iii and O25'-H25'N212iv; for symmetry codes and geometry see Table 2), while those to the water molecule (O100) employ different atoms as acceptors. For (I-1), atom O15' functions as acceptor (O100-H102O15'i), whereas atom O23' is the acceptor for (I-2) (O100-H101O23'). Additional weak contacts (Steiner, 2002) were observed for both conformers, including that of conformer (I-2) to atom O100 of the water molecule (C210-H210O100vi, C12-H12O13'v and C22-H22O23'vi).
| || Figure 1 |
Perspective views of (a) conformer (I-1) and (b) conformer (I-2), showing the atom-numbering schemes. Displacement ellipsoids are drawn at the 50% probability level and H atoms are shown as small spheres of arbitrary size.
| || Figure 2 |
Overlay of molecules (I-1) and (I-2).
| || Figure 3 |
The crystal packing showing the intermolecular hydrogen-bonding network (parallel to the ac plane).
Compound (I) was synthesized as reported previously (Seela et al., 2007). Slow crystallization from aqueous methanol afforded (I)·0.5H2O as colourless crystals (m.p. 442 K). For the diffraction experiment, a single crystal was mounted on a MiTeGen MicroMountsfibre in a thin smear of oil.
In the absence of suitable anomalous scattering, Friedel equivalents could not be used to determine the absolute structure. Refinement of the Flack (1983) parameter led to inconclusive values for this parameter [0.2 (7)]. Therefore, Friedel equivalents (2881) were merged before the final refinement. The known configuration of the parent molecule was used to define the enantiomer employed in the refined model.
All H atoms were found in a difference Fourier synthesis. In order to maximize the data/parameter ratio, the H atoms were placed in geometrically idealized positions (C-H = 0.95-1.00 Å) and constrained to ride on their parent atoms, with Uiso(H) = 1.2Ueq(C). The OH groups were refined as rigid groups, allowed to rotate but not tip (AFIX 147 instruction in the XL routine of SHELXTL; Sheldrick, 2008), with O-H = 0.84 Å and Uiso(H) = 1.5Ueq(O). The water H atoms were located from difference maps, and the parameters of the water H atoms were first refined freely. Owing to the low reflection/refined parameter ratio, the O-H distances were constrained [AFIX 3 (m = 0)] to 0.96 Å and with Uiso(H) = 1.5Ueq(O) in the final cycles of refinement.
Data collection: APEX2 (Bruker, 2008); cell refinement: SAINT (Bruker, 2008); data reduction: SAINT; program(s) used to solve structure: SHELXTL (Sheldrick, 2008); program(s) used to refine structure: SHELXTL; molecular graphics: DIAMOND (Brandenburg, 1999); software used to prepare material for publication: SHELXTL and PLATON (Spek, 2009).
Supplementary data for this paper are available from the IUCr electronic archives (Reference: GD3377 ). Services for accessing these data are described at the back of the journal.
Altona, C. & Sundaralingam, M. (1972). J. Am. Chem. Soc. 94, 8205-8212.
Arab, K., Pedersen, M., Nair, J., Meerang, M., Knudsen, L. E. & Bartsch, H. (2009). Carcinogenesis, 30, 282-285.
Bielecki, L., Skalski, B., Zagorowska, I., Verrall, R. E. & Adamiak, R. W. (2000). Nucleosides Nucleotides Nucleic Acids, 19, 1735-1750.
Bolt, H. M. (1994). DNA Adducts: Identification and Biological Significance, Vol. 125, edited by K. Hemminki, A. Dipple, D. E. G. Shuker, F. F. Kadlubar, D. Segerback & H. Bartsch, pp. 141-150. Lyon: IARC.
Brandenburg, K. (1999). DIAMOND. Crystal Impact GbR, Bonn, Germany.
Bruker (2008). APEX2 (Version 2008/5), SADABS (Version 2008/1) and SAINT (Version 7.56a). Bruker AXS Inc., Madison, Wisconsin, USA.
Flack, H. D. (1983). Acta Cryst. A39, 876-881.
Inoue, Y., Kuramochi, T. & Imakubo, K. (1981). Chem. Lett. 10, 1161-1164.
IUPAC-IUB Joint Commission on Biochemical Nomenclature (1983). Eur. J. Biochem. 131, 9-15.
Jaskólski, M. (1982). Acta Cryst. B38, 3171-3174.
Lin, W., Seela, F., Eickmeier, H. & Reuter, H. (2004). Acta Cryst. C60, o566-o568.
Pandya, G. A. & Moriya, M. (1996). Biochemistry, 35, 11487-11492.
Paulsen, H. & Wintermeyer, W. (1984). Eur. J. Biochem. 138, 125-130.
Secrist, J. A., Barrio, J. R., Leonard, N. J. & Weber, G. (1972). Biochemistry, 11, 3499-3506.
Seela, F., Schweinberger, E., Xu, K., Sirivolu, V. R., Rosemeyer, H. & Becker, E.-M. (2007). Tetrahedron, 63, 3471-3482.
Sheldrick, G. M. (2008). Acta Cryst. A64, 112-122.
Spek, A. L. (2009). Acta Cryst. D65, 148-155.
Steiner, T. (2002). Angew. Chem. Int. Ed. 41, 48-76.
Van Wijk, L., Haasnoot, C. A. G., de Leeuw, F. A. A. M., Huckriede, B. D., Westra Hoekzema, A. J. A. & Altona, C. (1999). PSEUROT6.3. Leiden Institute of Chemistry, Leiden University, The Netherlands.
Zabel, V., Saenger, W. & Seela, F. (1987). Acta Cryst. C43, 131-134.