ABSTRACT
The incorporation of 1-deazaadenosine (c1A, 1b) into a hammerhead ribozyme and the resulting catalytic activity is described. For this purpose the phosphoramidite 2a and the 3'-phosphonate 2b as well as Fractosil-linked 1-deazaadenosine (3b) were prepared. The methoxyacetyl group was used for the 6-amino group protection and the triisopropylsilyl residue was introduced as the 2'-OH protecting group. Replacement of residues A14 and A15.1 of the hammerhead ribozyme by 1-deazaadenosine resulted in a significantly reduced catalytic activity. Substitution of the A6, A9 and A13 residues has only a minor influence. The findings observed on ribozymes modified with 1-deazaadenosine were compared with those containing other adenosine analogues.
Hammerhead ribozymes are small self-cleaving RNA-molecules made up of three base-paired stems flanking a highly conserved central core. Cleavage of the RNA occurs as the result of a transesterification reaction and generates two products: one containing a terminal 5'-hydroxyl group and the second a terminal 2',3'-cyclic phosphate. Several crystal structures of hammerhead ribozymes have been published. Self cleavage was avoided by (i) methylation of the 2'-hydroxyl group necessary for catalytic activity (1), (ii) incorporation of a 2'-deoxyribonucleoside residue (2), (iii) the absence of divalent metal ions (3). The sequence-specific nature of this endoribonuclease activity suggests ribozymes may be therapeutic agents for the inhibition of gene expression (4).
Despite significant progress and a number of models, the detailed mechanism of the hammerhead ribozyme cleavage is still unknown. The substitution of individual nucleoside residues by nucleoside analogues having a specific base structure, a modified sugar or phosphate residue, has been used to probe the catalytic activity of individual functional groups. The adenosine analogues studied include: (i) purine (5,6), (ii) 7-deazaadenosine (tubercidin) (7,8) and (iii) 3-deazaadenosine (9). To determine the importance of specific adenosine nitrogen-1 [N(1)] for the catalytic action, adenosine was replaced by 1-deazaadenosine (c1A, 1b) (purine numbering is used throughout the discussion section) in the hammerhead ribozyme core (Fig. 1). By this means the potential N(1) proton acceptor site was removed.
The N(1) of adenosine can participate in a normal A-U Watson-Crick base pair and in a number of unusual inter- and intra-strand hydrogen bonding interactions. These include A-N(1) to G-N(1)H (imino) hydrogen bonds reported for G5-A26 of the high-affinity Rev binding site of the Rev responsive element of the HIV-1 mRNA (10). The same type of interaction was recently observed in a G-A mismatch of the model duplex (rGCGGACGC)2 (11). Intra-strand interaction of the GAAA tetraloop with an RNA helix was first described for the hammerhead system (2) where several adenosine residues form an A-N(1) hydrogen bond with the amino groups of another A- or G-residue and also with the 2'-hydroxyl group of G in the G-C stem. A similar hydrogen bonding was observed in the X-ray structure for the group 1 ribozyme domain (12) for the GAAA tetraloop interactions with an 11 nucleotide tetraloop receptor together with a symmetrical A-A [N(1)-NH2] base pair as a part of the reverse Hoogsteen A-AU base triple. A number of N(1)-A(2'-OH) hydrogen bonds were also observed in this structure.
In the crystal structures of the hammerhead ribozyme inhibitor complexes (1,2), only N(1) of the A14 residue is participating in a hydrogen bond; the N(1) acts as acceptor of the 2'-OH group of the non-conserved U7 residue. There is no indication based on the X-ray data that N(1) of the conserved adenosine residues A15.1, A13, A9 and A6 is participating in hydrogen bonds.
The 1-deazaadenosine substitution resulted in substantial reductions in the rate of hammerhead cleavage (under single-turnover conditions) when substituted at positions A15.1 and A14 (krel = 0.005, 0.004). This result suggests that the N(1) functional groups at these positions are important in the stabilization of the transition state or conformational intermediate immediately preceding the transition state. These results are not immediately consistent with the current crystal structure models (see discussion below). The A13, A9 and A6 substitutions resulted in more modest decreases in cleavage rates (krel = 0.03, 0.1 and 0.03, respectively). We suggest that these reductions in rate are likely the result of indirect effects on ribozyme structure.
1-Deazaadenosine (1b) was at first prepared by Chatterjee et al. by condensation of the chloromercury derivative of 7-acetamido-imidazo[4,5-b]pyridine with 2,3,5-tri-O-benzoyl-1-chloro-[beta]-d-ribofuranose in boiling xylene (13). This method furnishes N(7) and N(9) regioisomers and results in a low overall yield. Later, De Roos and Salemink synthesized compound 1b by the melting-fusion procedure in a slightly higher yield (14). Furthermore, Itoh et al. have described another synthesis of 1b starting from imidazo-[4,5-b]pyridine-4-oxide which has been transformed in five steps into the desired 1-deazaadenosine (1b) in a good overall yield (15,16). Recently, Cristalli et al. reported a more convenient synthetic route (17) via the SnCl4-catalyzed glycosylation of 7-nitroimidazo[4,5-b]pyridine with 1,2,3,5-tetra-O-acetyl-[beta]-d-ribofuranose (TAR), followed by the deprotection with methanolic ammonia and reduction of the 6-nitro group. 1-Deazaadenosine (1b), required for this work, was synthesized either according to Mizuno or to Cristalli (Scheme 1).
Scheme 1.
Table 1.
The pKa value of the protonation of 1-deazaadenosine (1b) is 4.7 (18). The protonation position of compound 1b was determined by 13C NMR spectroscopy and found to be N(3) (Table 1). This was derived from the upfield protonation shifts of the [alpha]-carbons next to N(3) while those in [beta]- and [gamma]-position are deshielded (Table 1) (19). The 1-deazaadenosine shows a UV spectrum (MeOH) with two long-wavelength absorption maxima at 264 nm ([epsilon] = 12 500) and 280 nm ([epsilon] = 10 100). The protonation site of the parent 2'-deoxy-adenosine determined by 15N NMR spectroscopy is N(1) (20). The pKa value of the protonation is 3.8 (21).
Oligoribonucleotides containing 1-deazaadenosine were prepared by solid-phase synthesis. For this purpose the building blocks 2a and 2b were synthesized (22,23). Earlier, it has been shown that the N-benzoyl group introduced in 1-deaza-2'-deoxyadenosine is a rather stable protecting group and can be removed only with difficulty (24). Therefore, the methoxyacetyl (mac) and phenoxyacetyl (pac) protecting groups were studied in this case. The mac derivative 4a was obtained via peracylation of compound 1b with methoxyacetyl chloride followed by selective deprotection of the sugar protecting groups with an excess of a pyridine-Et3N-H2O (1:1:3) (89% yield) (25). In a similar way, the pac derivative 4b was introduced with phenoxyacetyl anhydride (73% yield). The stability of the protecting groups was measured UV-spectrophotometrically. The half-life time ([tau]) of 4a (mac) is 17 min, whereas the pac derivative 4b gave a value of 11 min, both measured in 25% aqueous NH3 solution at 40°C (Table 2). The mac derivative 4a was used for further work. The 4,4'-dimethoxytritylation was introduced under standard conditions furnishing compound 5 (88% yield) (Scheme 2).
Scheme 2.
The five ribozymes with individual substitution at positions A15.1, A14, A13, A9 and A6 by 1-deazaadenosine (8-12) were tested in a cleavage assay containing the complementary 17mer substrate (Fig. 1) under single-turnover conditions. The parent (unmodified) complex of this previously characterized ribozyme (32) showed a cleavage rate of 0.52/min (kobs). The relative cleavage rates (krel) of the base-modified ribozymes are displayed in Figure 1. The catalytic rate of the ribozymes with substitutions at the residues A6, A13 and A9 was reduced 10-30-fold while the substitutions of A14 and A15.1 resulted in a >200-fold decrease in the cleavage rate. As a control experiment, when the non-conserved position Al2.3 (position within loop II) was substituted no significant decrease in activity was observed. Since the cleavage reactions were performed under single turnover conditions with a large excess of the ribozyme (1 µM = 1000 × KD = 10 × KM of the unmodified ribozyme) and a trace amount of substrate, the observed cleavage rates should approximate the rate of the chemical reaction step (k2). Previous kinetic analysis of the parent ribozyme demonstrated that at this riboyzme concentration the observed rates were log-linear with pH (slope = 1) and magnesium concentrations. Reactions were also performed at 5 µM ribozyme concentration and identical rates were observed, consistent with saturating concentrations of the ribozyme reaction (data not shown).
If the N(1) of the conserved adenosine residues is involved in hydrogen-bonding that is important for the stabilization of the transition state and/or conformational changes towards the transition state, the absence of this acceptor should result in a substantial decrease in k2. The magnitude of this decrease should be related to the free energy supplied by the hydrogen bond [[Delta]G = -RT·ln(k2)]. The free energy supplied by a hydrogen bond may vary but has been measured to be 0.2-1.5 kcal/mol. One would, therefore, expect a 1.2-5-fold decrease in k2. Based upon the above assumptions, a possible explanation for the replacement of adenosine by 1-deazaadenosine would be that each N(1) within the core of the hammerhead is involved in specific interactions, since the cleavage rate always dropped at least 10-fold. On the other hand, this simple explanation is not consistent with the X-ray data and requires further clarification.
There are two factors which should be taken into consideration in the analysis of the activity changes of the modified ribozymes. The first important consideration is that the current crystal structure model very likely represents the `ground state' of the ribozyme (33). The interactions between functional groups of the conserved nucleotides that are stabilizing the `ground state' may differ significantly from the interactions stabilizing the transition state. Therefore, functional groups that are seemingly unimportant in the ground state could have a significant effect in the transition state or accompanying conformational changes. In fact, several functional groups of that type have been identified in the catalytic core of the hammerhead domain through the `atomic mutagenesis' approach including all functional groups of the guanine at G5, as well as nitrogen-3 [N(3)] of the adenine A15.1-residues and also of others (33).
The second important factor that has to be considered in any atomic mutagenesis study relates to non-specific physicochemical changes of the nucleoside residue that result from even single atom substitution. These nonspecific effects can disrupt other interactions and thereby decrease activity. For example, the pKa value of N(3) is altered by the N(1) to C(1) modification. The pKa value of N(3)-protonation of adenosine is 3.5 (34) while that of N(3)-protonation of 1-deazaadenosine is 4.7 (18). This reflects the higher electron density of the 1-deazaadenine system compared with that of adenine. Additionally, the 6-amino group of 1-deazaadenosine is a more basic residue as in the case of adenosine, therefore acting as a weaker proton donor. This is reflected by the much higher stability of the 1-deazaadenosine amino protecting groups. However, because the cleavage reactions were performed at a pH well above the pKa value of the base (pH 6.5), one would not expect this effect to be substantial. The higher electron density of the 1-deazaadenine system compared with that of adenine could manifest itself in different stacking interactions of the base. Finally, 1-deazadenosine shows a larger preference for a syn-conformation around the glycosidic bond compared with adenosine (35,36).
For all of these reasons, it is very difficult to predict the exact magnitude of the decrease in k2 that results from any single atom modification. However, the ribozymes substituted by 1-deazaadenosine fall into two distinct classes [ <= 30-fold decrease (A6, A9, A13) versus >200-fold decrease (A14, A15.1)]. It is tempting to argue that the removal of the N(1) from A14 and A15.1 results from the loss of the H-acceptor site, whereas the removal of the N(1) from A6, A9 and A13 is due to other phenomena. It is also possible to better separate the specific from non-specific effects by comparing catalytic activities of ribozymes substituted by 1-deazaadenosine with previously characterized ribozymes substituted by other adenosine analogues.
It was reported earlier that 3-deazaadenosine and 7-deazaadenosine at position A14 results only in a 4- and 8-fold reduction of the cleavage rate (7,9). In addition, substitution of the A14 residue by purine riboside leads only to a 3-7-fold decrease of activity (5,6). If the large decrease of the ribozyme activity substituted at A14 with 1-deazaadenosine (250-fold) is the result of effects not related to hydrogen bonding, one would expect a significant decrease in activity by other base modifications at that position. To the best of our knowledge, the results obtained from ribozymes substituted by 1-deazaadenosine are the first indication that N(1) of A14 is important for the hammerhead function. In the X-ray structure this nitrogen (A14) participates in a hydrogen bond with the 2'-hydroxyl group of the uridine residue of U7. However, a 2'-deoxyuridine replacement at position U7 has essentially no effect on the catalytic rate as many other 2'-modifications (37). It is therefore very unlikely that the absence of this interaction in the 1-deazadenosine-A14 mutant results in the observed 250-fold drop in the cleavage activity.
Previous results obtained from the modification at position A15.1 are somewhat controversial. Fu and McLaughlin (6) reported a 6-fold decrease of the cleavage activity by the incorporation of the purine riboside (nebularine). Slim and Gait (5) found a substantial loss of the activity for this kind of replacement under different cleavage conditions. In our hands, under standard single-turnover conditions, the substitution of A15.1 by nebularine results in a 30-fold decrease of the cleavage activity (data not shown). The finding that the 1-deazaadenosine incorporation (A15.1) results in a much larger activity loss cannot be simply due to the decreased proton donor capability of the 6-amino group. Surprisingly, in two crystal structures (1,2) an interaction of A15.1 with U16.1 was found. A single hydrogen bond is formed between the 6-amino group and O(4). Ruffner and Uhlenbeck (38) demonstrated that any substitution of the A15.1-U16.1 base pair by e.g. a U-A, C-G or G-C pair resulted in complete loss of activity. The incorporation of 7-deazaadenosine at position A15.1 resulted in a small enhancement of the cleavage activity (7), while the removal of N(3) (3-deazaadenosine) resulted in a large decrease in cleavage rate (9). Taken together the results argue for direct participation of N(1) and N(3) at position A15.1.
The presence of nitrogen-7 at position A6 was reported to be important for the hammerhead cleavage activity. This was the result of the incorporation of 7-deazaadenosine leading to a 200-fold reduction (8). On the other hand the incorporation of purine as well as 3-deazaadenosine at the same position leads only to a 1.5- and 4-fold decrease of the activity. This indicates the relative unimportance of these residues (4,5,9). The 30-fold reduction observed after 1-deazaadenosine incorporation falls in between the effects caused by 7-deazaadenosine and 3-deazaadenosine. In the crystal structure, adenosine at A6 occupies a central position in domain 1 or the `uridine turn' forming a purine platform A6-G5 near the cleavage site (1,2). The adenine ring of A6 is involved in perpendicular stacking interactions with the furanose oxygen-4' of the C17 residue in the cleavage site with N(1) of A6 in close proximity to O(4') of C17. The G5 residue is a well recognized anomaly in attempts to reconcile the results of functional group modifications with the X-ray structures because a large number of modifications at G5 inhibit hammerhead activity, yet this residue does not interact with any other functional groups of the ribozyme in the published crystal structures. We (39) and others suggested that there is a need for a substantial conformational change with possible rearrangement of domain 1 and especially G5 in a transition state compared with the ground state of the hammerhead-substrate complex. On the other hand, based on structures of rapidly frozen ribozyme-substrate complexes, Scott et al. (3) argued that small conformational adjustments are sufficient for achieving catalytic conformation. An additional argument for a substantial rearrangement of domain 1 in the hammerhead upon cleavage was recently presented by Simore et al. (40), based on NMR analysis on the cleaved hammerhead-product complex, where formation of new hydrogen bonds between U4 and U7 was observed.
Several factors may contribute to a 30-fold decrease in cleavage activity for 1-deazaadenosine substitution at position A6: the altered electron density of the adenine ring in this analog can effect the formation of the A6-G5 purine platform and/or on an aromatic interaction of the adenine ring of A6 and the furanose oxygen of the C17 residue at the cleavage site. If we also assume that substantial rearrangement in domain 1 (C3-U7) takes place in the transition state, then such a rearrangement will involve the base of A6 because of its central location. In such a scenario the altered pKa value of the 6-amino group of 1-deazaadenosine can account for the formation of unproductive destabilizing interactions.
At position A9 the 1-deazadenosine substitution caused the smallest effect (10-fold decrease) in cleavage activity which indicates the relative unimportance of N(1) at this position. Furthermore, it was demonstrated that the 6-amino group, N(3) and N(7) functionalities also have only a modest impact on cleavage activity based on purine (2-fold reduction) 3-deazaadenosine (10-fold reduction ) and 7-deazaadenosine (2-fold reduction) substitutions (4-9). This is surprising since an abasic ribose moiety at this conserved position resulted in a >2500-fold reduction in cleavage activity which can be recovered almost completely by adding adenine base in trans, indicating the perfect fit of adenine functionalities (41).
According to the crystal structures, the adenosine residue at position A13 is directly involved in the formation of `GA-shared' base pairs G12A13-A9G8. Additional evidence for the formation of these shared G-A base pairs in the solution was recently obtained by NMR spectroscopy (37). The essential part of these interactions is the stacking of A9 above A13. It was reported (4-9) that 7- and 3-deazaadenosine and purine substitutions at this position resulted in moderate (2-, 8- and 8-fold, respectively) decreases of the cleavage activity. The observed 30-fold decrease in cleavage activity for 1-deazaadenosine substitution at this position probably reflects more drastic changes in the physical properties of this analog. At the same time, the effect does not seem to be large enough to justify direct involvement of N(1) in stabilization of the vital interaction in the transition state. It is more likely that the altered electron density of the purine ring of 1-deazaadenosine destabilizes stacking interactions which are essential for providing correct geometry of the GA-shared pairs G12A13-A9G8 in catalytically productive intermediate.
One of the most powerful and common tools for understanding structure-function relationships of small RNA molecules is the single atom mutagenesis approach. This technique relies upon individually replacing a single atom or functional group within the entire RNA molecule, and then determining the function of that RNA molecule. A single atom mutation is certainly the smallest and most specific mutation possible and one would therefore expect that changes in the function of the RNA could be correlated directly to the atom or functional group that was mutated. Unfortunately, by mutating a single atom many different features of the nucleotide and RNA may change. For example, the properties of adjacent functional groups may be altered (pKa), the stacking properties (charge distribution, electron density, hydrophobicity) of the base may change, the syn/anti equilibrium about the glycosidic bond can be effected, and finally conformation of the sugar can shift. It is therefore necessary to carefully compare the effect of substituting several different positions and examine the effects of other modifications at the same position to interpret the results.
To elucidate the importance of the N(1) nitrogens of the conserved adenosine residues for hammerhead cleavage, we prepared the phosphoramidite and phosphonate building blocks (2a,2b) of 1-deazaadenosine protected with a methoxyacetyl group at the amino function and the triisopropylsilyl group at the 2'-OH. When this modified nucleoside was placed individually at each of the five conserved adenosine residues, activity always dropped by at least 10-fold. The simplest explanation for this result is that every A-N(1) within the core of the hammerhead ribozyme is involved in a critical interaction. This would be particularly surprising since current structure models and biochemical evidence argue that none of the A-N(1) nitrogens are important for activity. Upon closer examination, however, it is clear that the 1-deazaadenosine substitutions within the core result in two classes of ribozymes: (i) >200-fold decreases in kobs (A14 and A15.1) and (ii) 10-30-fold decreases in kobs (A6, A9 and A13). Substitution of position Al2.3 outside of the core resulted in <2-fold decrease in kobs. We suggest that all of the 1-deazaadenosine-substituted ribozymes suffer decreases in activity due to indirect changes to the nucleoside. However, because the magnitude of the decrease in kobs for A14- and A15.1-substituted ribozymes and the results with other ribozymes containing adenosine analogues, we argue that the N(1) nitrogens of A14 and A15.1 are involved in critical interactions for stabilizing the transition state or conformational changes immediately preceding the transition state.
TLC. Alumina sheets coated with a 0.2 mm layer of silica gel 60 F254 (Merck, Germany). Flash chromatography (FC) was carried out at 0.5 bar.
UV Spectra. Hitachi U-3000 and U-3200 spectrophotometer (Hitachi, Japan).
NMR Spectra. Bruker AC-250 and AMX-500 spectrometers; [delta] values in p.p.m. relative to tetramethylsilane as internal standard (1H and 13C) or to external phosphoric acid (31P). The J values are in Hz. The elemental analyses were performed by Mikroanalytisches Laboratorium Beller (Göttingen, Germany).
Theoligoribonucleotides (2.5 µmol scale) were synthesized on polystyrene supports (ABI) on an Applied Biosystems 394 DNA/RNA synthesizer. Synthesis, deprotection and analysis of the oligoribonucleotides were performed as previously described (30). The ribozymes were purified on 15% polyacrylamide, TBE (89 mM Tris-borate, 2 mM EDTA) and 7 M urea gels. The full-length RNA was identified by UV shadowing of the gel, passively eluted with 10 mM Tris-HCl (pH 8.0) and 1 mM EDTA, and then concentrated by ethanol precipitation. The final ribozyme concentrations were determined by the UV absorbance using an extinction coefficient of 350/mM/cm at 260 nm.
To verify the incorporation of 1-deazaadenosine, aliquots of purified ribozymes were analyzed by nucleoside composition analysis (34). Briefly, the modified ribozymes were digested to the ribonucleosides by incubation of 0.3 A260 units of the oligonucleotide with 10 U of nuclease P1 (Boehringer Mannheim) and 2 U of calf intestinal alkaline phosphatase (Boehringer Mannheim) in 30 mM NaOAc and 1 mM ZnSO4, at pH 5.2 (total volume = 100 µl) overnight at 50°C. The digested material was injected directly onto a C18 reverse column (Waters, Symmetry, ODS 4.6 × 250 mm), and the ribonucleotides were separated by a gradient of buffers A (50 mM potassium phosphate, pH 7.0) and B (95% aqueous MeOH); 0-90% B in A over 35 min. The retention times were compared with monomer standards which were independently characterized by NMR. Under these conditions 1-deazaadenosine was eluted with a retention time of 12 min (Fig. 2).
Ribozyme oligonucleotides and 5'-32P-end-labeled substrates were heated separately in the reaction buffer at 95°C for 1 min, quenched on ice and then equilibrated to the final reaction temperature (37°C) prior to starting reactions. Reactions were carried out with an excess of the enzyme and were initiated by mixing equal volumes (20 µl) of the substrate (final concentration <1 nM) and ribozyme (1 µM) in 50 mM MES pH 6.5, 10 mM MgCl2. Aliquots (4 µl) were removed at various times (5 s to 2 h), quenched in 8 µl of formamide loading buffer (95% formamide, 20 mM EDTA) and loaded onto 15% polyacrylamide, TBE (89 mM Tris-borate, 2 mM EDTA) gels containing 7 M urea. The fraction of substrate and product present at each time point was determined by quantitation of scanned images using a Molecular Dynamics PhosphorImager. The ribozyme cleavage rates were obtained from the plots of the fraction of the substrate remaining versus time using a non-linear, least-squares fit to a double exponential curve (KaleidaGraph, Synergy Software, Reading, PA). The initial, fast portion of the curve represented 80-90% of the total reaction; thus the observed cleavage rates (kobs) were taken from the rate constant for the first exponential. Relative rates of cleavage (krel) were calculated by dividing the observed cleavage rate by the cleavage rate of the unmodified ribozyme (0.52/min). The total extent of cleavage was always >70%.
1-Deazaadenosine (1b) (15-17) (410 mg, 1.5 mmol) was co-evaporated several times with anhydrous pyridine (20 ml, each), dissolved in dry pyridine (20 ml) and treated with methoxyacetyl chloride (1.1 ml, 12 mmol). After 1.5 h of stirring at room temperature, H2O (3 ml) was added to destroy the excess methoxyacetyl chloride. After another 30 min, the solution was evaporated, the residue dissolved in CH2Cl2 (60 ml) and extracted three times with 5% aqueous NaHCO3 solution followed by H2O (30 ml). The organic layer was dried (Na2SO4) and evaporated. The oily residue was stirred for 45 min with Et3N-pyridine-H2O (1:1:3) (70 ml) at room temperature, the solution evaporated, the oily residue was co-evaporated twice with toluene (20 ml, each) and the residue applied to FC (silica gel, column 15×3 cm, CH2Cl2-MeOH 9:1). The migrating zone was evaporated to give colorless crystals of 4a from acetone (450 mg, 89%); melting point 160-162°C [Found: C, 49.38; H, 5.44; N, 16.51. Calculated for C14H18N4O6 (338.3): C, 49.70; H, 5.36; N, 16.56%]; TLC (silica gel; 15×3 cm, CH2Cl2-MeOH 9:1) 0.30; [lambda]max (MeOH/nm) 271 and 281 ([epsilon]/dm3/mol/cm 20 800 and 17 600); [delta]H [500 MHz; (d)6DMSO] 3.43 (3 H, s, OMe), 3.97 (2 H, m, 5'-H2), 3.97 (1 H, m, 4'-H), 4.17 (1 H, m, 3'-H), 4.17 (2 H, s, CH2), 4.66 (1 H, m, 2'-H), 5.21 (1 H, d, J 4.3, 3'-OH), 5.34 (1 H, t, J 5.4, 5'-OH), 5.48 (1 H, d, J 5.6, 2'-OH), 6.10 (1 H, d, J 5.9, 1'-H), 8.09 (1 H, d, J 5.4, 6-H), 8.27 (1 H, d, J 5.4, 5-H), 8.63 (1 H, s, 2-H), 9.68 (1 H, s, NH).
Compound 1b (180 mg, 0.68 mmol) was dried by evaporation (twice) with anhydrous pyridine (10 ml), dissolved in dry pyridine (10 ml), and treated with phenoxyacetic anhydride (1.56 g, 5.45 mmol). After 12 h of stirring at room temperature, H2O (2 ml) was added to destroy the anhydride. After another 30 min, the solution was evaporated, the residue was dissolved in CH2Cl2 (10 ml) and extracted three times with 5% aqueous NaHCO3 followed by H2O (50 ml each). The organic layer was dried (Na2SO4) and evaporated to dryness. The resulting yellow oil was stirred for 30 min with pyridine-Et3N-H2O (1:1:3) (40 ml) at room temperature, the solution was evaporated, co-evaporated twice with toluene (20 ml) and the residue was applied to FC (silica gel, 15 × 3 cm, CH2Cl2-MeOH 9:1). The migrating zone was evaporated to yield colorless crystals from MeOH (200 mg, 73%), melting point 202-204°C (Found: C, 56.85; H, 5.96; N, 13.91. Calculated for C19H20N4O6 (400.4): C, 57.00; H, 5.03; N, 13.99%); TLC (silica gel; CH2Cl2-MeOH 9:1) 0.38; [lambda]max (MeOH/nm) 265sh, 272, and 282sh ([epsilon]/dm3/mol/cm 18 500, 21 100, and 17 300); [delta]H [250 MHz; (d6)DMSO] 3.65 (2 H, m, 5'-H2), 3.99 (1 H, m, 4'-H), 4.19 (1 H, m, 3'-H), 4.67 (1 H, m, 2'-H), 4.94 (2 H, s, CH2), 5.19 (1 H, d, J 4.6, 3'-OH), 5.32 (1 H, t, J 5.5, 5'-OH), 5.46 (1 H, d, J 6.1 2'-OH), 6.04 (1 H, d, J 5.8, 1'-H), 6.94-7.35 (5 H, m, ArH), 8.11 (1 H, d, J 5.5, 6-H), 8.27 (1 H, d, J 5.5, 5-H), 8.66 (1 H, s, 2-H), 10.5 (1 H, s, NH).
Compound 4a (400 mg, 1.2 mmol) was dried by repeated co-evaporation with anhydrous pyridine and dissolved in dry pyridine (5 ml). Then 4,4'-dimethoxytrityl chloride (520 mg, 1.5 mmol) was added and the solution stirred for 3 h under argon at 40°C. The mixture was cooled to room temperature, MeOH (3 ml) was added and the mixture stirred for another 1 h. Then, the solution was reduced to half of the volume, dissolved in CH2Cl2 (50 ml), and extracted three times with 5% aqueous NaHCO3 solution (20 ml) followed by saturated aqueous NaCl solution (20 ml). The organic layer was dried (Na2SO4), filtered and evaporated. The residue was purified by FC (silica gel, 15 × 3 cm; CH2Cl2-MeOH 95:5) to yield a colorless foam (670 mg, 88%), (Found: C, 65.41; H, 5.77; N, 8.75. Calculated for C35H36N4O8 (640.7): C, 65.61; H, 5.66; N, 8.74%); TLC (silica gel; CH2Cl2-MeOH 95:5) 0.40; [lambda]max (MeOH/nm) 234, 271 and, 280 ([epsilon]/dm3/mol/cm 22 100, 22 300 and 19 000); [delta]H [250 MHz; (d6)DMSO] 3.22 (2 H, m, 5'-H2), 3.43 (3 H, s, OMe), 3.71 (3 H, s, OMe), 4.09 (1 H, m, 4'-H), 4.18 (2 H, s, CH2), 4.34 (1 H, m, 3'-H), 4.79 (1 H, m, 2'-H), 5.25 (1 H, d, J 5.4, 3'-OH), 5.58 (1 H, d, J 5.4, 2'-OH), 6.07 (1 H, d, J 3.9, 1'-H), 6.78-7.37 (13 H, m, ArH), 8.10 (1 H, d, J 5.3, 6-H), 5.28 (1 H, d, J 5.3, 5-H), 8.54 (1 H, s, 2-H), 9.87 (1 H, s, NH).
Method A. A solution of 5 (315 mg, 0.49 mmol) in dry pyridine (4 ml) was cooled to 0°C, treated with triisopropylsilyl chloride (Pri3SiCl) (130 µl, 0.64 mmol) and 1H-imidazole (67 mg, 0.98 mmol), stirred for 1 h at 0°C and then for 23 h at room temperature, then a second portion of 1H-imidazole (33 mg, 0.47 mmol) and Pri3SiCl (98 µl, 0.49 mmol) was added. The mixture was stirred for 24 h at room temperature. Then 5% aqueous NaHCO3 solution (20 ml) was added, the solution was extracted with CH2Cl2 (3 × 20 ml), the combined organic layers were dried (Na2SO4), filtered and evaporated. The residue was applied to FC (silica gel, 25 × 3 cm, petrolether-EtOAc-CH2Cl2 1:1:1). From the faster migration zone, 6 (195 mg, 50%) was isolated as a colorless foam.
Method B. To a solution of compound 5 (420 mg, 0.66 mmol) in dry pyridine (5 ml), AgNO3 (170 mg, 1.00 mmol) was added. After 5 min, a solution of Pri3SiCl (150 µl, 0.75 mmol) in dry THF (7 ml) was introduced under exclusion of light and moisture. After 24 h of stirring, a second portion of Pri3SiCl (75 µl, 0.38 mmol) and AgNO3 (55 mg, 0.32 mmol) was added, and stirring was continued for another 12 h. AgCl was filtered off and the filtrate treated with 5% aqueous NaHCO3 solution (20 ml). The aqueous layer was extracted (three times) with CH2Cl2 (20 ml). The combined organic layers were washed with H2O (30 ml) and saturated aqueous NaCl solution (30 ml), dried (Na2SO4) and evaporated to dryness. The residue was applied to FC (silica gel, 25 × 3 cm, petrolether-EtOAc-CH2Cl2 1:1:1). Compound 6 was isolated from the faster migrating main zone as a colorless foam (400 mg, 76%); (Found: C, 66.61; H, 7.06; N, 7.12. Calculated for C44H56N4O8Si (797.1): C, 66.31; H, 7.08; N, 7.03%); TLC (silica gel; petrolether-EtOAc-CH2Cl2 1:1:1) 0.8; [lambda]max (MeOH/nm) 235, 271 and 281 ([epsilon]/dm3/mol/cm 23 100, 23 500, and 19 600); [delta]H [250 MHz; (d6)DMSO] 0.78-0.89 (21 H, m, Pri3Si), 3.28 (2 H, m, 5'-H2), 3.43 (3 H, s, OMe), 3.71 (6 H, s, OMe), 4.17 (1 H, m, 4'-H), 4.17 (2 H, s, CH2), 4.31 (1 H, m, 3'-H), 5.15 (1 H, m, 2'-H), 5.17 (1 H, m, 3'-OH), 6.11 (1 H, d, J 5.4, 1'-H), 6.61-7.41 (13 H, m, ArH), 8.09 (1 H, d, J 5.4, 6-H), 8.18 (1 H, d, J 5.4, 5-H), 8.56 (1 H, s, 2-H), 9.86 (1 H, s, NH).
From the slower migrating zone the 3'-O-silylated derivative 7 was obtained as a colorless foam (A: 80 mg, 20%; B: 75 mg, 14%). (Found: C, 66.19; H, 7.00; N, 7.21. Calculated for C44H56N4O8Si (797.1): C, 66.31; H, 7.08; N, 7.03%); TLC (silica gel; petrolether-EtOAc-CH2Cl2 1:1:1) 0.5; [lambda]max (MeOH/nm) 235, 270, and 280 ([epsilon]/dm3/mol/cm 22 500, 22 700, and 19 100); [delta]H [500 MHz; (d6)DMSO] 1.01-1.04 (21 H, m, Pri3Si), 3.28 (2 H, m, 5'-H2), 3.46 (3 H, s, OMe), 3.73 (6 H, s, OMe), 4.13 (1 H, m, 4'-H), 4.19 (2 H, s, CH2), 4.70 (1 H, m, 3'-H), 5.00 (1 H, m, 2'-H), 5.39 (1 H, d, J 5.4, 2'-OH), 6.06 (1 H, d, J 4.9, 1'-H), 6.80-7.35 (13 H, m, ArH), 8.12 (1 H, d, J 5.3, 6-H), 8.21 (1 H, d, J 5.3, 5-H), 8.59 (1 H, s, 2-H), 9.83 (1 H, s, NH).
Diisopropylethylamine (460 µl; 2.65 mmol) and (2-cyanoethyl)(N,N-diisopropylamino)-phosphite (230 µl; 1.03 mmol) were added to a flask containing dry THF (5 ml) under argon (syringe). Then, a solution of 6 (400 mg; 0.50 mmol) in dry THF (5 ml) was added, and the reaction mixture was stirred overnight at room temperature. The reaction was monitored by TLC. The reaction mixture was diluted with CH2Cl2 (100 ml) and washed with 5% aqueous NaHCO3 solution (50 ml) followed by saturated brine (50 ml). The organic layer was dried (Na2SO4), evaporated and co-evaporated with dry acetone. The residue was dissolved in diethyl ether. Evaporation yielded the diastereomeric mixture of the building block 2a (410 mg; 82%) as a colorless foam; TLC (silica gel; CH2Cl2-n-hexane-Et3N 6:3:1) 0.5; C53H73N6O9SiP (997.3), [delta]P [101 MHz; CDCl3] 149.2, 152.2.
To a solution of PCl3 (290 µl, 3.4 mmol) and N-methylmorpholine (3.8 ml, 32.2 mmol) in anhydrous CH2Cl2 (30 ml), 1H-1,2,4-triazole (770 mg, 11.1 mmol) was added and the mixture stirred for 30 min at room temperature. After cooling to 0°C, a solution of compound 6 (530 mg, 0.66 mmol) in dry CH2Cl2 (20 ml) was added dropwise within 10 min. The mixture was stirred for 30 min at room temperature, and poured into 1 M aqueous (Et3NH)HCO3 solution (TBK buffer, 30 ml pH 7.5). The aqueous solution was extracted with CH2Cl2 (3 × 15 ml), the combined organic layer dried with Na2SO4, filtered and evaporated. FC [silica gel, 15 × 2 cm, CH2Cl2-Et3N 98:2 (300 ml), then CH2Cl2-MeOH-Et3N 88:10:2] furnished a colorless residue. It was dissolved in CH2Cl2, washed with 0.1 M aqueous (Et3NH)HCO3 (5 × 25 ml), dried with Na2SO4, and co-evaporation with acetone furnished a colorless foam (570 mg, 90%); [Found: C, 62.30; H, 7.55; N, 7.21. Calculated for C50H72N5O10PSi (962.2): C, 62.41; H, 7.54; N, 7.28%]; TLC (silica gel; CH2Cl2-MeOH-Et3N 88:10:2) 0.7; [lambda]max (MeOH/nm) 235, 271, and 281 ([epsilon]/dm3/mol/cm 22 400, 23 300 and 19 400); [delta]H [250 MHz; (d6)DMSO] 0.72-0.85 (21 H, m, Pri3Si), 1.12 (9 H, t, J 7.2, (CH3CH2)3NH), 2.89 (6 H, q, J 6.9, (CH3CH2)3NH), 3.31 (2 H, m, 5'-H2), 3.42 (3 H, s, OMe), 3.72 (6 H, s, OMe), 4.18 (H, s, CH2), 4.40 (1 H, m, 4'-H), 4.68 (1 H, m, 3'-H), 5.33 (1 H, m, 2'-H), 6.12 (1 H, d, J 6.9, 1'-H), 6.76 (1 H, d, J 594, PH), 6.80-7.88 (13 H, m, ArH), 8.07 (1 H, d, J 5.4, 6-H), 8.12 (1 H, d, J 5.4, 5-H), 8.07 (1 H, d, J 5.4, 6-H), 8.12 (1 H, d, J 5.4, 5-H), 8.45 (1 H, s, 2-H), 9.88 (1 H, s, NH); [delta]p [101 MHz; (CD3)2SO] 2.28 (1J(31P,H) 594, 3J(31P,3'-H) 9.3.
To a solution of compound 7 (210 mg, 0.26 mmol) in anhydrous 1,2-dichloroethane (0.6 ml), 4-(N,N-dimethylamino)pyridine (16 mg, 0.13 mmol), succinic anhydride (40 mg, 0.53 mmol) and Et3N (36 µl, 0.26 mmol) were added. The mixture was stirred for 30 min at 50°C. Then the solution was diluted with 1,2-dichloroethane (10 ml) and washed with ice-cold citric acid solution (3 × 10 ml) and H2O (3 × 10 ml). The organic layer was dried with Na2SO4, filtered, evaporated and the residue submitted to FC (silica gel, 15 × 3 cm, acetonitrile-H2O 9:1). The main zone yielded a colourless powder (185 mg, 79%); (Found: C, 64.34; H, 6.89; N, 6.32. Calculated for C48H60N4O11Si (897.1): C, 64.27; H, 6.74; N, 6.25%); TLC (silica gel; acetonitrile-H2O 9:1) 0.8; [lambda]max (MeOH/nm) 236, 270, and 280 ([epsilon]/dm3/mol/cm 22 100, 22 600 and 18 800); [delta]H [500 MHz; (d6)DMSO] 0.96-0.98 (21 H, m, Pri3Si), 2.45-2.67 (4 H, m, (CH2)2), 3.27 (2 H, m, 5'-H2), 3.45 (3 H, s, OMe), 7.71 (6 H, s, OMe), 4.12 (1 H, m, 4'-H), 4.18 (2 H, s, CH2), 5.37 (1 H, m, 3'-H), 6.15 (1 H, m, 2'-H), 6.26 (1 H, m, 1'-H), 6.73-7.26 (13 H, m, ArH), 8.17 (1 H, d, J 5.4, 6-H), 8.28 (1 H, d, J 5.4, 5-H), 8.60 (1 H, s, 2-H), 9.85 (1 H, s, NH).
A solution of 3a (120 mg; 0.13 mmol) in 1,4-dioxane containing 5% pyridine (1 ml) was treated with 4-nitrophenol (33 mg; 0.24 mmol) and N,N'-dicyclohexylcarbodiimide (50 mg; 0.24 mmol). The mixture was stirred for 4 h. Then, N,N'-dicyclohexylurea was removed by filtration. The filtrate was added to a suspension of amino-linked silica gel (200 mg Fractosil-200 450 mmol/NH2/g; Merck) in dry DMF (2 ml) and Et3N (200 ml). After shaking overnight, Ac2O (60 ml) was added and shaking was continued for another 30 min. The silica gel was filtered off, washed with DMF, EtOH and absolute Et2O, and dried in vacuum. The amount of silica-gel-bound nucleoside was determined by treatment of 3b (5 mg) with 0.1 M TsOH (10 ml) in acetonitrile. The amount of the Fractosil linked nucleoside was found to be 93 µmol/ g [[epsilon]498(DMT) = 71 000].
We thank Dr H. Rosemeyer for measurements of the NMR spectra, V. Mokler for the synthesis of the ribozymes and S. Newman for base composition analysis. We also thank Dr F. Wincott for critical reading of the manuscript. Financial support by the Deutsche Forschungsgemeinschaft (DFG) is gratefully acknowledged.
Nucleic Acids Research
Pages
Introduction
Results And Discussion
Monomer and oligoribonucleotide synthesis
Cleavage activity of the ribozymes modified by 1-deazaadenosine
Comparison of the cleavage activity of the c1A-modified ribozymes with other base-modified ribozyme molecules
Conclusion
Materials And Methods
General
Oligonucleotide synthesis
Substrate cleavage assay
7-(Methoxyacetylamino)-3-([beta]-d-ribofuranosyl)-3H-imidazo-[4,5-b]pyridine (4a)
7-(Phenoxyacetylamino)-3-([beta]-d-ribofuranosyl)-[4,5-b]-pyridine (4b)
3-[5-O-(4,4'-Dimethoxytriphenylmethyl)-[beta]-d-ribofuranosyl]-7-(methoxyacetylamino)-3H-imidazo[4,5-b]pyridine (5)
3-{5-O-(4,4'-Dimethoxytriphenylmethyl)-2-O-[tris(1-methyl-ethyl)silyl]-[beta]-d-ribofuranosyl}-7-(methoxyacetylamino)-imidazo[4,5-b]pyridine (6)
3-{5-O-(4,4'-Dimethoxytriphenylmethyl)-3-O-[tris(1-methyl-ethyl)silyl]-[beta]-d-ribofuranosyl}-7-(methoxyacetylamino)-3H-imidazo[4,5-b]pyridine (7)
3-{5-O-(4,4'-Dimethoxytriphenylmethyl)-2-O-[tris(1-methyl-ethyl)silyl]-[beta]-d-ribofuranosyl}-7-(methoxyacetylamino)-3H-imidazo[4,5-b]pyridine 3'-[(2-cyanoethyl) N,N-diisopropyl-phosphoramidite] (2a)
3-{5-O-(4,4'-Dimethoxytriphenylmethyl)-2-O-[tris(1-methyl-ethyl)silyl]-[beta]-d-ribofuranosyl}-7-(methoxyacetylamino)-3H-imidazo[4,5-b]pyridine 3'-triethylammonium phosphonate (2b)
3-{5-O-(4,4'-Dimethoxytriphenylmethyl)-2-O-[tris(1-methyl-ethyl)silyl]-[beta]-d-ribofuranosyl}-7-(methoxyacetylamino)-3H-imidazo[4,5-b]pyridine 3'-(3-carboxypropanoate) (3a)
3-{5-O-(4,4'-Dimethoxytriphenylmethyl)-2-O-[tris(1-methyl-ethyl)silyl]-[beta]-d-ribofuranosyl}-7-(methoxyacetylamino)-3H-imidazo[4,5-b]pyridine 3'-[3-(N-fractosilcarbamoyl)-propanoate] (3b)
Acknowledgements
References
a
C(1)
C(2)
C(4)
C(5)
C(6)
C(7)
Neutral
102.4
144.2
146.5
123.8
147.4
140.0
Acidicb
103.1
136.8
138.7
123.2
152.0
140.0
REFERENCES
This page is run by Oxford University Press, Great Clarendon Street, Oxford OX2 6DP, as part of the OUP Journals
Comments and feedback: www-admin{at}oup.co.uk
Last modification: 6 Feb 1998
Copyright© Oxford University Press, 1998.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  J. A. Nelson and O. C. Uhlenbeck Hammerhead redux: Does the new structure fit the old biochemical data? RNA, April 1, 2008; 14(4): 605 - 615. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Kojima, K. Inoue, R. Nakajima-Shibata, S.-i. Kawahara, and E. Ohtsuka A new, but old, nucleoside analog: the first synthesis of 1-deaza-2'-deoxyguanosine and its properties as a nucleoside and as oligodeoxynucleotides Nucleic Acids Res., December 15, 2003; 31(24): 7175 - 7188. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. B. Tuite, J. C. Shiels, and A. M. Baranger Substitution of an essential adenine in the U1A-RNA complex with a non-polar isostere Nucleic Acids Res., December 1, 2002; 30(23): 5269 - 5275. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. C. Shiels, J. B. Tuite, S. J. Nolan, and A. M. Baranger Investigation of a conserved stacking interaction in target site recognition by the U1A protein Nucleic Acids Res., January 15, 2002; 30(2): 550 - 558. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Takagi, M. Warashina, W. J. Stec, K. Yoshinari, and K. Taira SURVEY AND SUMMARY: Recent advances in the elucidation of the mechanisms of action of ribozymes Nucleic Acids Res., May 1, 2001; 29(9): 1815 - 1834. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S.A. STROBEL, G.W. MUTH, and L. CHEN Exploring the Mechanism of the Peptidyl Transfer Reaction by Chemical Footprinting Cold Spring Harb Symp Quant Biol, January 1, 2001; 66(0): 109 - 118. [Abstract] [PDF]
|
|
|
|
 |
|
 G. W. Muth, L. Ortoleva-Donnelly, and S. A. Strobel A Single Adenosine with a Neutral pKa in the Ribosomal Peptidyl Transferase Center Science, August 11, 2000; 289(5481): 947 - 950. [Abstract] [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||