6-Pyrazolylpurine as an Artificial Nucleobase for Metal-Mediated Base Pairing in DNA Duplexes.

The artificial nucleobase 6-pyrazol-1-yl-purine (6PP) has been investigated with respect to its usability in metal-mediated base pairing. As was shown by temperature-dependent UV spectroscopy, 6PP may form weakly stabilizing 6PP-Ag(I)-6PP homo base pairs. Interestingly, 6PP can be used to selectively recognize a complementary pyrimidine nucleobase. The addition of Ag(I) to a DNA duplex comprising a central 6PP:C mispair (C = cytosine) leads to a slight destabilization of the duplex. In contrast, a stabilizing 6PP-Ag(I)-T base pair is formed with a complementary thymine (T) residue. It is interesting to note that 6PP is capable of differentiating between the pyrimidine moieties despite the fact that it is not as sterically crowded as 6-(3,5-dimethylpyrazol-1-yl)purine, an artificial nucleobase that had previously been suggested for the recognition of nucleic acid sequences via the formation of a metal-mediated base pair. Hence, the additional methyl groups of 6-(3,5-dimethylpyrazol-1-yl)purine may not be required for the specific recognition of the complementary nucleobase.

1. Introduction

Nucleic acids are increasingly being used as building blocks in nanotechnology, owing to their superb and highly predictable self-assembly, their stiffness, and the ease of their modification [1]. An important way to introduce class="Chemical">metal-based functionality into nucleic acids is the use of class="Chemical">pan class="Chemical">metal-mediated base pairs. In these artificial base pairs, coordinative bonds to one or two central metal ions formally replace the hydrogen bonds usually found between complementary nucleobases [2,3,4,5]. As confirmed by several structural studies, the metal ions align along the helical axis inside B-DNA duplexes [6,7,8]. Similarly, Z-DNA and A-RNA duplexes are also compatible with the formation of metal-mediated base pairs [9,10,11]. The resulting metal-containing nucleic acids have found manifold applications, including their use as sensors [12], in expanding the genetic code [13], in modifying the charge transfer capabilities of DNA [14,15], in the generation of metal nanoclusters [16], and in the recognition of other nucleic acid sequences [17].

In the latter context, a series of class="Chemical">purine derivatives with aclass="Chemical">pclass="Chemical">pended class="Chemical">pan class="Species">donor moieties have recently been proposed as artificial nucleobases for the specific recognition of a complementary canonical nucleobase via the formation of a metal-mediated base pair [18,19,20,21]. For one of these derivatives, i.e., 6-(3,5-dimethylpyrazol-1-yl)purine, single-crystal X-ray diffraction analysis of a model nucleobase confirmed the suggested binding pattern of the transition metal ions, namely, via the pyrazole nitrogen atom and the purine N7 position [22]. The same study also indicated that this binding pattern is retained when formally removing the methyl groups from the pyrazole rings, i.e., when using 6-pyrazol-1-yl-purine. Moreover, due to the lack of a steric clash between the (absent) methyl groups, the model nucleobase 9-methyl-6-pyrazol-1-yl-purine was shown to be able to form homoleptic 2:1 complexes with Cu(II) and Ag(I) [22]. As this artificial purine derivative had never been tested as a nucleobase for metal-mediated base pairing inside a DNA duplex, we set out to investigate its corresponding properties. This study reports that 6-pyrazolyl-1-yl-purine (6PP) is able to form an Ag(I)-mediated base pair with a complementary thymine residue, but that stabilizing homo base pairs of the type 6PP–M+–6PP are not formed inside a DNA duplex.

2. Results

2.1. Synthesis of the Deoxyribonucleoside and the DNA Duplex

The artificial nucleobase class="Chemical">6-pyrazol-1-yl-purine 1 was synthesized by stirring class="Chemical">pan class="Chemical">6-chloropurine in neat pyrazole at 150 °C (Scheme 1) [23]. Reaction of 1 with Hoffer’s chloro sugar gave the β isomer of the p-toluoyl-protected deoxyribonucleoside 2. Finally, deprotection by means of aqueous methanolic ammonia resulted in the formation of the desired 6PP deoxyribonucleoside 3. To enable its use in automated DNA oligonucleotide solid-phase synthesis, the hydroxyl groups were protected in two subsequent steps with 4,4′-dimethoxytrityl (5′-OH) and 2-cyanoethyl-N,N-diisopropyl-phosphoramidite (3′-OH) to give the 6PP phosphoramidite building block 5.

Scheme 1

Synthesis of the 6-pyrazolyl-1-yl-purine (6PP) deoxyribonucleoside (3) via free 6PP (1) and p-toluoyl-protected 6PP deoxyribonucleoside (2). a) Neat, 150 °C, 1 h; b) first step: NaH (1.6 equiv.), CH3CN, 0 °C, 1 h, second step: Hoffer’s chloro sugar, toluene, overnight; c) aqueous NH3, methanol, RT, 4 h.

The class="Chemical">oligonucleotide duclass="Chemical">plex investigated in this study is shown in Scheme 2. It was chosen because several other class="Chemical">pan class="Chemical">metal-mediated base pairs have already been characterized in this sequence context, thereby enabling a direct comparison with other systems. The duplex comprises 12 natural base pairs plus one central artificial base pair (denoted as X:Y in Scheme 2).

Scheme 2

Oligonucleotide duplex under investigation.

2.2. Spectroscopic Characterization of the Metal-Containing DNA Duplex

2.2.1. Homo Base Pair (6PP:6PP)

As the structural studies using the model nucleobase class="Chemical">9-methyl-6-pyrazol-1-yl-purine had suggested the class="Chemical">possibility of the formation of a class="Chemical">pan class="Chemical">metal-mediated homo base pair from 6PP [22], we initially investigated the incorporation of metal ions into a DNA duplex comprising a central 6PP:6PP mispair. The formation of the anticipated metal-mediated base pair was examined by determining the melting temperatures of the duplex Tm in the presence of increasing amounts of suitable metal ions. As the formation of coordinate bonds between the artificial nucleobases and the metal ion typically leads to an increased stability of the duplex [2], monitoring the Tm often represents a good indication for the formation of a metal-mediated base pair. Exceptions from this rule are rare, but are sometimes found in the case of non-planar metal-mediated base pairs [24,25] or when the metal-free duplex is exceptionally stable [26]. Figure 1 shows the melting curves of the duplex comprising a 6PP:6PP base pair in the absence and presence of one equivalent of AgNO3. Prior to the addition of AgNO3, the duplex melts at 34.8 °C. This melting temperature is in the upper range of melting temperatures reported for the same sequence with other artificial nucleobases (21.7–39.1 °C) [27] but is significantly below that of a duplex comprising a central A:T or G:C base pair (42.5, 45.4 °C) [28]. In the presence of one Ag(I) per duplex, Tm increases by 2.5 °C to a final value of 37.3 °C. This rather small increase does not unambiguously confirm the formation of a metal-mediated base pair. However, considering that previous studies showed that a small thermal stabilization does not rule out the formation of a metal-mediated base pair [24,25], in particular when the base pair is non-planar, no final conclusion regarding the formation of a 6PP–Ag(I)–6PP base pair can be drawn.

Figure 1

UV melting curves of the duplex comprising a central 6PP:6PP base pair in the absence (black) and presence (red) of one equivalent of AgNO3.

A putative class="Chemical">6PP–Ag(I)–class="Chemical">pan class="Chemical">6PP base pair could exist with coordinate bonds involving the purine N1 atom, i.e., via the Watson–Crick edge (Scheme 3a), or the purine N7 atom, i.e., via the Hoogsteen edge (Scheme 3b). Both types of metal-mediated base pairing patterns have previously been reported for other purine derivatives within antiparallel-stranded duplexes [29,30,31]. Considering that all metal complexes of the model nucleobase showed coordination via N7 [22], this binding pattern (Scheme 3b) is also the more likely one for the putative 6PP–Ag(I)–6PP base pair.

Scheme 3

Possible structures of the putative 6PP–Ag(I)–6PP base pair (R, R′ = DNA backbone). (a) Watson–Crick-type binding via N1; (b) Hoogsteen-type binding via N7.

Attempts to introduce other class="Chemical">metal ions led to an even smaller change in the case of class="Chemical">pan class="Chemical">Cu(II) (∆Tm = +1.0 °C) and no change at all for Ni(II), Hg(II), and Zn(II). For the latter three metal ions, the formation of a metal-mediated base pair can be excluded. For Ni(II), this is surprising, because the closely related 6-pyridylpurine had been reported to form exceptionally stabilizing Ni(II)-mediated homo base pairs (∆Tm = 18 °C) [30].

2.2.2. Hetero Base Pairs (6PP:C and 6PP:T)

Following the initial notion that class="Chemical">6-(3,5-dimethylpyrazol-1-yl)-purine may be able to sclass="Chemical">pecifically recognize a canonical nucleobase [18], we also investigated this class="Chemical">possibility with the sterically less demanding class="Chemical">pan class="Chemical">6-pyrazol-1-yl-purine 6PP. As 6PP is a purine derivative, we limited our investigation to the base pairing with the pyrimidine nucleobases cytosine and thymine. Moreover, as only Ag(I) had a significant influence on the melting temperature of the duplex comprising 6PP:6PP, we focused on the use of AgNO3 in the study of the possible metal-mediated 6PP:C and 6PP:T base pairs. Figure 2 shows the melting curves of the duplexes containing these artificial base pairs in their center. As thymine is protonated at N3, the study involving the thymine-containing base pair was performed at an elevated pH in order to facilitate deprotonation and subsequent coordination to Ag(I).

Figure 2

UV melting curves of the duplex comprising a central 6PP:pyrimidine base pair in the presence of increasing amounts of AgNO3 (black: 0 equiv.; red: 1. equiv.; blue: 2 equiv.; orange: 3 equiv.) (a) 6PP:C (pH 6.8); (b) 6PP:T (pH 9.0). The arrow indicates the direction of the changes upon the addition of AgNO3. The inset shows the increase in Tm depending on the amount of added AgNO3.

In the absence of Ag(I), both duplexes melt at around 29 °C. As expected, the duplex is slightly destabilized at pH 9.0 with respect to the one at pH 6.8 (28.1 °C vs. 30.1 °C). The average melting temperature is significantly lower than that of the duplex with a central class="Chemical">6PP:6PP base class="Chemical">pair and reflects the imclass="Chemical">portance of π stacking interactions in stabilizing a nucleic acid duclass="Chemical">plex. Aclass="Chemical">pclass="Chemical">parently, the reduced π surface of the class="Chemical">pan class="Chemical">pyrimidine nucleobases compared with the purine derivative effect this decrease in stability.

It is interesting to note that the addition of class="Chemical">AgNO3 to the duclass="Chemical">plex containing a central class="Chemical">pan class="Chemical">6PP:C base pair (Figure 2a) slightly destabilizes the duplex even further (∆Tm = −2 °C). In contrast, when Ag(I) is added to the duplex with a central 6PP:T base pair, a significant increase in Tm is observed (∆Tm = 4.3 °C). Moreover, the addition of excess AgNO3 does not significantly influence the melting temperature anymore. This is a clear indication of the formation of a mononuclear Ag(I)-mediated base pair, as the extra stability is conferred by the incorporation of the first Ag(I) only [2]. Similarly, a plot of the circular dichroism (CD) spectra of the duplex comprising the central 6PP:T base pair in the presence of increasing amounts of AgNO3 shows negligible changes during the stepwise addition of the first equivalent but significant changes in the presence of excess Ag(I) (Figure 3). This indicates that the overall duplex structure is not disrupted by the formation of the 6PP–Ag(I)–T base pair and that excess Ag(I) also binds to duplex, albeit at other positions [28,32].

Figure 3

Circular dichroism (CD) spectra of the duplex comprising a central 6PP:T base pair in the presence of increasing amounts of AgNO3 (0, 0.25, 0.5, 0.75, 1, 1.5, 2, 3 equiv.). The arrow indicates the direction of the changes upon the addition of AgNO3.

Scheme 4 shows possible structures of the class="Chemical">6PP–Ag(I)–T base class="Chemical">pair. Similarly to class="Chemical">pan class="Chemical">6PP–Ag(I)–6PP, the Ag(I) ion may bind via N1 or via N7. In both cases, a [2+1] coordination environment is proposed for the metal ion, in analogy to what has been found in other Ag(I)-mediated base pairs [26,33]. Weak additional binding via one of the carbonyl oxygen atoms cannot be ruled out though. As the crystal structures of the complexes of the corresponding model nucleobase had shown that N7 is the preferred metal binding site of 6PP, the arrangement shown in Scheme 4b appears to be more likely. In this structure, the possible formation of a C–H···O hydrogen bond involving an aromatic hydrogen atom of the pyrazole moiety may even reinforce the preference of the metal ion for N7.

Scheme 4

Possible structures of the 6PP–Ag(I)–T base pair (R, R′ = DNA backbone). (a) Watson–Crick-type binding via N1; (b) Hoogsteen-type binding via N7.

3. Discussion

The data presented here show that class="Chemical">6PP reclass="Chemical">presents a valuable addition to the list of artificial class="Chemical">pan class="Chemical">nucleobases for metal-mediated base pairing. Out of the several substituted purine derivatives published to date [18,19,28,30,31,34,35,36,37], it adds to those ones with an N,N-donor set [18,30]. Its putative metal-mediated homo base pairs do not evoke a strong thermal stabilization of the DNA duplex. In contrast, it readily forms a stabilizing Ag(I)-mediated base pair with the canonical nucleobase thymine. Even though it is not as sterically crowded as its derivative 6-(3,5-dimethylpyrazol-1-yl)purine, 6PP has the potential to discriminate between a complementary cytosine and a thymine via the formation of an Ag(I)-mediated base pair. A similar, albeit not as pronounced, stability trend has been reported for the metal-mediated base pairs formed from 6-(3,5-dimethylpyrazol-1-yl)purine and the pyrimidine nucleobases in 2′-O-methyl RNA oligonucleotides in the presence of Cu(II) [18]. The similarities between 6-(3,5-dimethylpyrazol-1-yl)purine and 6-pyrazol-1-yl-purine suggest that the methyl substituents on the pyrazole moiety may not be required to achieve the specific recognition of a complementary nucleobase.

4. Materials and Methods

4.1. Oligonucleotide Synthesis and Characterization

class="Chemical">Phosclass="Chemical">phoramidites required for the synthesis of the class="Chemical">pan class="Chemical">oligonucleotide sequences were from Glen Research. Syntheses of the oligonucleotide strands were performed on a K & A Laborgeräte H8 DNA/RNA synthesizer in the DMT-off mode by following standard protocols. Post-synthesis, the oligonucleotides were cleaved from the solid support and deprotected by treating them with tert-butylamine/MeOH/NH3 (1/2/1) for 4 h at 60 °C. Thereafter, they were purified by denaturing urea polyacrylamide gel electrophoresis (gel solution: 7 M urea, 1 M TBE buffer, 18% polyacrylamide:bisacrylamide (29:1); loading buffer: 11.8 M urea, 42 mM Tris/HCl (pH 7.5), 0.83 mM EDTA (pH 8.0), 8% sucrose, 0.08% dye (xylene cyanol, bromophenol blue)). After purification, the oligonucleotides were desalted with NAP10 columns. The desalted oligonucleotides were characterized by MALDI-TOF mass spectrometry (5′-d(GAG GGA XAG AAA G)-3′: calcd. for [M + H]+: 4157 Da, found: 4156 Da; 5′-d(CTC CCT XTC TTT C)-3′: calcd. for [M + H]+: 3863 Da, found: 3862 Da; 5′-d(CTC CCT TTC TTT C)-3′: calcd. for [M + H]+: 3803 Da, found: 3802 Da; 5′-d(CTC CCT CTC TTT C)-3′: calcd. for [M + H]+: 3789 Da, found: 3787 Da). MALDI-TOF mass spectra were recorded on a Bruker Reflex IV instrument using a 3-hydroxypicolinic acid/ammonium citrate matrix. High resolution (positive mode) mass spectra were obtained on a Waters Q-Tof Premier Micromass HAB 213 mass spectrometer or on an LTQ Orbitrap XL (Thermo Scientific, Bremen, Germany), equipped with the static nanospray probe (slightly modified to use self-drawn glass nanospray capillaries, spray voltage 1.0–1.4 kV, capillary temperature 200 °C, tube lens approx. 50–120 V). During the quantification of the oligonucleotides, a molar extinction coefficient ε260 of 4.8 cm2·μmol−1 was used for 6PP deoxyribonucleoside.

4.2. Spectroscopy

NMR spectra were recorded using Bruker Avance(I) 400 and Bruker Avance(III) 400 spectrometers. Chemical shifts were recorded with reference to residual solvent peak in class="Chemical">DMSO-d6 (class="Chemical">pan class="Chemical">1H NMR δ = 2.49 ppm, 13C NMR δ = 39.5 ppm), CDCl3 (1H NMR δ = 7.26 ppm, 13C NMR δ = 77.2 ppm), CD3CN (1H NMR δ = 1.94 ppm, 13C NMR δ = 118.3 ppm), or with respect to trimethylsilyl propane sulfonate (D2O, δ = 0 ppm). UV melting experiments were carried out on a UV spectrometer CARY 100 Bio using a 1-cm quartz cuvette. The UV melting profiles were measured at 260 nm in buffer at pH 6.8 (1 μM oligonucleotide duplex, 150 mM NaClO4, 5 mM MOPS) or pH 9.0 (1 μM oligonucleotide duplex, 150 mM NaClO4, 5 mM borate), either in absence or in presence of AgNO3, at a heating rate of 1 °C·min–1 with data being recorded at an interval of 0.5 °C. Prior to each measurement, the sample was equilibrated by heating it to 60 °C followed by cooling to 10 °C at a rate of 1 °C·min–1. Melting temperatures were determined from the maxima of the first derivatives of the melting curves.

4.3. 6-(1H-Pyrazol-1-yl)-9H-purine 1

Compound 1 was synthesized according to a literature procedure [23]. Its class="Chemical">1H NMR sclass="Chemical">pectrum agreed well with the literature data. Additional characterization data are as follows. class="Chemical">pan class="Chemical">13C NMR (101 MHz, DMSO-d6), δ: 151.1 (C2p), 146.8 (C4p), 144.2 (C8p), 138.0 (C6p), 130.9 (C5p), 127.7 (C3*), 115.4 (C5*), 109.2 (C4*) ppm. ESI-MS m/z: 187.0727 [M + H]+ (calcd. 187.0732), [M + Na]+ 209.0546 (calcd. 209.0552). Elemental analysis (%): found: C 51.6, H 3.3, N 44.7 calcd. for C8H6N6: C 51.6, H 3.3, N 45.1.

4.4. 9-(2-Deoxy-3,5-di-O-p-toluoyl-β-d-erythro-pentofuranosyl)-6-(1H-pyrazol-1-yl)-9H-purine 2

class="Chemical">6-(1H-Pyrazol-1-yl)-9H-purine 1 (0.500 g, 2.69 mmol) was susclass="Chemical">pended in dry class="Chemical">pan class="Chemical">acetonitrile (20 mL), and NaH (0.170 g of a 60% suspension on oil, 4.26 mmol) was added. The resulting mixture was stirred for 1 h under ice cooling, followed by the addition of Hoffer’s chloro sugar [38] (1.25 g, 3.22 mmol), which was previously suspended in dry toluene (20 mL), over a period of 20 min in four portions. The solution was stirred overnight and the solvent evaporated. The oily residue was purified by silica gel column chromatography (cyclohexane (8):CH2Cl2 (1):EtOAc (4):Et3N (3)) to produce an off-white foam of the desired β anomer 2. Formation of the undesired α anomer was not observed. Yield: 0.673 g (48%). 1H NMR (400 MHz, CDCl3), δ: 9.00 (d, 1H, H5*), 8.75 (s, 1H, H2p), 8.49 (s, 1H, H8p), 7.97 (d, 1H, H3*), 7.93 (d, 2H, o-H (Tol)), 7.58 (d, 2H, o-H (Tol)), 7.25 (d, 2H, m-H (Tol)), 7.10 (d, 2H, m-H (Tol)), 6.71 (dd, 1H, H1′), 6.56 (s, 1H, H4*), 5.70 (m, 1H, H3′), 4.92 (m, 1H, H4′), 4.62 (m, 2H, H5′/H5′′), 3.14 (m, 1H, H2′ or H2′′), 3.09 (m, 1H, H2′ or H2′′), 2.40 (s, 3H, CH3), 2.31 (s, 3H, CH3) ppm. 13C NMR (101 MHz, CDCl3), δ: 166.0 (C=O), 165.7 (C=O), 153.2 (C4p), 152.0 (C2p), 147.2 (C6p), 144.5 (Tol), 144.2 (Tol), 144.6 (C3*), 142.6 (C8p), 130.6 (C5*), 129.6 (Tol), 129.3 (Tol), 129.2 (Tol), 129.2 (Tol), 126.6 (Tol), 125.8 (Tol), 122.6 (C5p), 108.9 (C4*), 86.1 (C1′), 84.7 (C4′), 74.9 (C3′), 63.9 (C5′), 38.4 (C2′), 21.6 (2 × CH3) ppm. ESI-MS m/z: [M + H]+ 539.2037 (calcd. 539.2043), [M + Na]+ 561.1857 (calcd. 561.1862). Elemental analysis (%): found: C 64.9, H 5.5, N 14.9; calcd. for C29H26N6O5·0.2 C6H12: C 65.3, H 5.2, N 15.1.

4.5. 9-(2-Deoxy-β-d-erythro-pentofuranosyl)-6-(1H-pyrazol-1-yl)-9H-purine 3

A solution of aqueous class="Chemical">NH3 (25%, 30 mL) in class="Chemical">pan class="Chemical">CH3OH (30 mL) was added to compound 2 (0.200 g, 0.37 mmol). After stirring for 4 h, the solvent was removed in vacuo, and the residue was purified by silica gel column chromatography (cyclohexane (8):EtOAc (5):CH2Cl2 (3):MeOH (1) → cyclohexane (8):EtOAc (5):CH2Cl2 (3):MeOH (3)) to produce a brown foamy solid of the free nucleoside 3. Yield: 0.107 g (95%). 1H NMR (400 MHz, D2O, pD 7.2) δ: 8.64 (s, 1H, H5*), 8.57 (s, 1H, H2p), 8.55 (d, 1H, H3*), 7.94 (s, 1H, H8p), 6.66 (m, 1H, H4*), 6.49 (dd, 1H, H1′), 4.69 (m, 1H, H3′), 4.18 (m, 1H, H4′), 3.85 (m, 2H, H5′/H5′′), 2.83 (m, 1H, H2′), 2.64 (m, 1H, H2′′) ppm. 13C NMR (101 MHz, D2O, pD 7.2), δ: 152.6 (C4p), 151.4 (C2p), 146.3 (C6p), 145.0 (C3*), 144.7 (C8p), 131.3 (C5*), 121.8 (C5p), 109.9 (C4*), 87.5 (C4′), 84.8 (C1′), 71.0 (C3′), 61.5 (C5′), 39.1 (C2′) ppm. ESI-MS m/z: [M + Na]+ 325.1020 (calcd. 325.1025). Elemental analysis (%): found: C 50.4, H 4.4, N 27.0; calcd. for C13H14N6O3·0.5H2O: C 50.2, H 4.9, N 27.0.

4.6. 9-[2-Deoxy-5-O-(4,4′-dimethoxytriphenylmethyl)-β-d-erythro-pentofuranosyl]-6-(1H-pyrazol-1-yl)-9H-purine 4

Compound 3 (106 mg, 0.334 mmol) was co-evaporated with dry class="Chemical">pyridine (2 × 20 mL) and dissolved in class="Chemical">pan class="Chemical">pyridine (10 mL) under argon atmosphere. After an addition of catalytic amounts of dimethylaminopyridine and 4,4′-dimethoxytrityl chloride (193 mg, 0.568 mmol), the mixture was stirred for 5 h. The solvent was removed in vacuo, and the crude material purified by silica gel column chromatography (CH2Cl2 (100):MeOH (1) → CH2Cl2 (95):MeOH (5)). After recrystallization from acetonitrile, product 4 was obtained as a white solid. Yield: 109 mg (54%). 1H NMR (400 MHz, CDCl3), δ: 9.05 (s, 1H, H5*), 8.73 (s, 1H, H2p), 8.29 (s, 1H, H8p), 7.96 (d, 1H, H3*) ,7.39 (m, 2H, DMT), 7.28 (m, 7H, DMT), 6.78 (m, 4H, DMT), 6.56 (m, 2H, H3*, H1′), 4.71 (m, 1H, H3′), 4.21 (m, 1H, H4′), 3.75 (s, 6H, CH3 (DMT)) 3.42 (m, 2H, H5′/H5′′), 2.87 (m, 1H, H2′), 2.62 (m, 1H, H2′′) ppm. 13C NMR (101 MHz, CDCl3), δ: 158.5 (DMT), 153.3 (C4p), 152.0 (C2p), 147.2 (C6p), 144.5 (C3*), 144.4 (DMT), 143.1 (C8p), 135.6 (DMT), 130.9 (C5*), 130.0 (DMT), 128.1 (DMT), 127.9 (DMT), 126.9 (DMT), 122.7 (C5p), 113.1 (DMT), 109.0 (C4*), 86.6 (DMT), 86.3 (C4′), 84.7 (C1′), 72.3 (C3′), 63.7 (C5′), 55.3 (DMT), 40.5 (C2′) ppm. ESI-MS m/z: [M + Na]+ 627.2363 (calcd. 627.2326). Elemental analysis (%): found: C 67.3, H 5.7, N 14.4; calcd. for C34H32N6O5: C 67.5, H 5.3, N 13.9.

4.7. 9-[2-Deoxy-5-O-(4,4′-dimethoxytriphenylmethyl)-β-d-erythro-pentofuranosyl]-6-(1H-pyrazol-1-yl)-9H-purine 3′-(2-cyanoethyl)-N,N′-diisopropyl phosphoramidite 5

The class="Chemical">DMT-class="Chemical">protected class="Chemical">pan class="Chemical">nucleoside 4 (162 mg, 0.268 mmol) was dissolved in dry CH2Cl2 (3 mL) under argon atmosphere. After adding N,N-diisopropylethylamine (0.238 mL, 0.113 mmol) and 2-cyanoethyl-N,N-diisopropylchlorophosporamidite (0.115 mL, 0.405 mmol), the mixture was stirred at ambient temperature for 2 h. The solvent was removed and the crude product purified by silica column chromatography (CH2Cl2 (75):EtOAc (23):NEt3 (2)). Yield 104 mg (48%). 1H NMR (400 MHz, CD3CN), δ: 9.12 (s, 1H, H5*), 8.69 (s, 1H, C2p), 8.40 (s, 1H, H8p), 7.90 (s, 1H, H3*), 7.36 (m, 2H, DMT), 7.23 (m, 7H, DMT), 6.75 (m, 4H, DMT), 6.63 (m, 1H, C4*), 6.51 (m, 1H, H1′), 4.91 (m, 1H, H3′), 4.25 (m, 1H, H4′), 3.82 (m, 2H, O–CH2), 3.71 (s, 6H, O–CH3), 3.62 (m, 2H, iPr-CH), 3.31 (m, 2H, H5′, H5′′), 3.08 (m, 1H, H2′), 2.65 (m, 3H, H2′′, CH2–CN), 1.19 (m, 12H, iPr-CH3) ppm. 13C NMR (101 MHz, CD3CN), δ: 159.7 (DMT), 154,5 (C4p), 152.7 (C2p), 148.2 (C6p), 145.5 (C3*), 144.7 (C8p), 136.8 (DMT), 132.5 (C5*), 131.1 (DMT), 131.0 (DMT), 129.0 (DMT), 127.8 (DMT), 124.0 (C5p), 118.3 (CN), 114.0 (DMT), 110.0 (C4*), 87.1 (DMT), 86.4 (C4′), 85.8 (C1′), 73.9 (C3′), 64.6 (C5′), 59.5 (O–CH2), 44.2 (iPr-CH), 39.3 (C2′), 24.9 (iPr-CH3), 21.1 (CH2–CN) ppm. 31P NMR (162 MHz, CD3CN), δ: 148.2, 148.1 ppm.