Synthesis of RNA Containing O-?-? ?-Ribofuranosyl-(1?? ??? ?2? ?)-adenosine-5?? ??-
phosphate and 1-Methyladenosine, Minor Components of tRNA
by Sergey N. Mikhailova), Ekaterina V. Efimtsevaa), Andrei A. Rodionova), Alexandra A. Shelkunovaa),
Jef Rozenskib), Gert Emmerechtsb), Guy Schepersb), Arthur Van Aerschotb), and Piet Herdewijn*b)
a) Engelhardt Institute of Molecular Biology, Russian Academy of Sciences, Vavilov str. 32, Moscow, 119991,
Russia (phone: ?7-095-1359704; fax: ?7-095-1351405; e-mail: email@example.com)
b) Rega Institute, Katholieke Universiteit Leuven, Minderbroedersstraat 10, B-3000 Leuven
(phone: ?32-16-337387; fax: ?32-16337340; e-mail: firstname.lastname@example.org)
tRNA is best known for its function as amino acid carrier in the translation process, using the anticodon
loop in the recognition process with mRNA. However, the impact of tRNA on cell function is much wider, and
mutations in tRNA can lead to a broad range of diseases. Although the cloverleaf structure of tRNA is well-
known based on X-ray-diffraction studies, little is known about the dynamics of this fold, the way structural
dynamics of tRNA is influenced by the modified nucleotides present in tRNA, and their influence on the
recognition of tRNA by synthetases, ribosomes, and other biomolecules. One of the reasons for this is the lack of
good synthetic methods to incorporate modified nucleotides in tRNA so that larger amounts become available
for NMR studies. Except of 2?-O-methylated nucleosides, only one other sugar-modified nucleoside is present in
tRNA, i.e., 2?-O-?-?-ribofuranosyl nucleosides. The T loop of tRNA often contains charged modified
nucleosides, of which 1-methyladenosine and phosphorylated disaccharide nucleosides are striking examples.
A protecting-group strategy was developed to introduce 1-methyladenosine and 5??-O-phoshorylated 2?-O-(?-?-
ribofuranosyl)-?-?-ribofuranosyladenine in the same RNA fragment. The phosphorylation of the disaccharide
nucleoside was performed after the assembly of the RNA on solid support. The modified RNA was
characterized by mass-spectrometry analysis from the RNase T1 digestion fragments. The successful synthesis of
this T loop of the tRNA of Schizosaccharomyces pombe initiator tRNAMetwill be followed by its structural
analysis by NMR and by studies on the influence of these modified nucleotides on dynamic interactions within
the complete tRNA.
Introduction. ± Nucleic acids consist of over 100 modified nucleosides,in addition to
eight major ribo- and deoxyribonucleosides. Although a few modifications have been
found in DNA, 86 modified nucleosides were found in tRNAs, and 15 were found in
other RNAs . What is the meaning of these modifications, and why does Nature use
so many of them? These general questions have not yet been answered in detail. Until
recently, few biochemical reactions and interactions of nucleic acids were known for
which a modified nucleoside was absolutely required . The modification of the
nucleosides is a posttranscriptional issue, and the determination of the function of
modified nucleosides has remained a challenge, which surpasses conventional research
approaches. Experimental approaches to the functions of modified nucleosides have
been limited to a comparison of completely modified RNA, purified from wild-type
organisms, to the corresponding mutants or completely unmodified RNA, which may
be prepared in large quantities, using in vitro T7 polymerase transcription of chemically
synthesized DNA templates or cloned genes . The importance of all these
modifications have multiple meanings, the significance of which is only fully under-
CHEMISTRY & BIODIVERSITY ± Vol. 2 (2005)1153
¹ 2005 Verlag Helvetica Chimica Acta AG, Z¸rich
stood when the contribution of one particular modification on the rest is investigated.
To get insight into structural and functional roles of modified nucleosides, effective
approaches for the preparation of tRNA should be further developed.
The first total synthesis of the biologically active yeast alanine tRNA containing
nine modified ribonucleosides was achieved in China in 1981 after 13 years of
collaboration of six laboratories  by the phosphodiester approach in solution. The
development of phosphoramidite chemistry resulted in an automated synthesis of
whole-unmodified tRNA . Later, chemical synthesis of modified tRNA containing
−simple× modifications (dihydrouridine, ribothymidine, and pseudouridine) was re-
ported . The enzymatic ligation of chemically synthesized RNA fragments is an
alternative approach in the preparation of tRNA [7±9]. The developed techniques
allow the synthesis of long oligonucleotides for biological investigation, but the
preparation of modified RNA in substantial quantity (1±3 mg) for structural studies
(NMR spectroscopy and X-ray analysis) is still a challenge for organic chemistry.
Most of the minor ribonucleosides in tRNA have modified heterocyclic bases. Only
two types of sugar-modified nucleosides were found in RNA, namely 2?-O-methyl and
2?-O-?-?-ribofuranosylderivatives. The latter,O-?-?-ribofuranosyl-(1???2?)-guanosine-
and adenosine-5??-phosphate (Grp and Arp, resp.), were isolated from initiator tRNAs
of yeasts and some plants [10±13]. The presence of a minor disaccharide nucleoside in
position 64 is suggested to be a common property of initiator tRNA of lower
eukaryotes. It was shown that this modification controlled the discrimination between
their initiator vs. elongator function . It should be mentioned that the charged
components are usually located in tRNA loops, but Grp and Arp are located in position
64 of the stem region of the T loop . The biosynthetic pathway of modified
nucleotides is still obscure. However, there is a hypothesis about posttranscriptional
ribosylation of adenosine or guanosine residues located in position 64 of the
cytoplasmic initiator tRNAMetprecursor with 5-phosphoribosyl-1-?-pyrophosphate to
form a ?-glycoside bond .
An X-ray analysis showed that a bulky phosphoribofuranosyl hydrophilic
substituent of Arp is strictly fixed on the surface of the tRNAMetminor groove. The
extra phosphate residue is involved in the formation of a H-bond with the 2-amino
group of the adjacent guanosine residue .
Another modification which frequently occurs in the T loop of tRNAs is 1-
methyladenosine (m1A). Approximately 25% of all tRNAs have m1A at position 58 in
the T loop, while m1A also often occurs at position 14 in the D loop . 1-
Methyladenosine is involved in the cloverleaf folding of tRNA , and its
chemical incorporation in RNA has been described . Here, we describe the
incorporation of Arp and m1A in the same RNA hairpin (T loop), which is largely an
exercise in protecting-group compatibilities under the conditions of RNA synthesis and
Results and Discussion. ± Recently, we have developed a general method for the
preparation of 2?-O-?-?-ribofuranosyladenosine, and studied its incorporation into
oligonucleotides (for reviews, see [19±21]). Independently, similar research was
carried out by Markiewicz et al. . It was shown that the thermal stabilities (at 15?) of
oligo(r(A)13¥r(U)13) and the fully modified oligo(r(A?)13¥r(U)13) are equal,
CHEMISTRY & BIODIVERSITY ± Vol. 2 (2005)1154
implying that the extra sugar moieties have no effect on the stability of the RNA duplex
(A??2?-O-?-?-ribofuranosyladenosine moiety) . The solution structure of the self-
complementary oligoribonucleotide sequence 5?-r(GCGA?AUUCGC) prepared from
synthon I was determined by high-resolution NMR spectroscopy and restrained
molecular dynamics . The bulky extra sugar does not disturb the A-type RNA
duplex and takes up a well-defined position in the minor groove, with its 5?-OH group
pointing to NH2?C(2)of theneighboring 5?G residue, in a very similar manner as in the
case of the above-mentioned crystal structure of tRNAMet.
Here, we first investigated the synthesis of the Arp phosphoramidite (Fig. 1) and its
successful incorporation into RNA for the first time. To achieve this goal, two strategies
have been considered: the phosphate may be introduced on the monomer level as in
synthon II, or by phosphorylation after oligonucleotide assembly on polymer support.
In the case of synthon III, the 5??-O-protecting group (X) should be selectively cleaved
to allow phosphorylation, followed by deblocking of the oligomer. In both cases, all of
the protecting groups should be compatible with disaccharide nucleoside and
oligonucleotide syntheses. The selective blocking of five OH groups in 2?-O-?-?-
ribofuranosylnucleosides is a specific challenge in the preparation of the title
The synthon of type II (with the guanine base) was synthesized starting from the
previously prepared compound 1 . The 2-(4-nitrophenyl)ethyl (npe) group
 was chosen for protection of the phosphate residue. Transformation of 1 to 3
was accomplished in good overall yields (Scheme 1). Synthesis of the RNA fragment
using 3 worked out fine according to trityl analysis during the assembly cycle, but
incomplete deprotection of the phosphate moiety of the ribose glycoside turned out to
be the bottleneck of this strategy. According to mass spectrometry, only one npe group
was cleaved with 1,8-diazabicyclo[5.4.0]undec-7-ene (DBU) or even stronger bases. It
was shown previously that the deprotection of some triester derivatives of 5?-
nucleotides with 0.5? DBU in pyridine was incomplete after five days .
Fig. 1. Structures of disaccharide nucleotides occuring in tRNA and protected monomers for incorporation in
CHEMISTRY & BIODIVERSITY ± Vol. 2 (2005) 1155
In view of several reports on selective deprotection of levulinoyl moieties in the
presence of other ester moieties [28±30], we decided to prepare synthon III (X?
levulinate), based on adenosine. Its synthesis is depicted in Scheme 1 starting from
1,2,3-tri-O-benzoyl-?-ribofuranose  readily available in three steps from ?-ribose.
Acylation in the presence of N,N-dicyclohexylcarbodiimide (DCC) resulted in
compound 4 in 91% yield. Ribosylation of N6,3?,5?-O-protected adenosine 5 with a
small excess of 4 in the presence of SnCl4under standard conditions  gave
disaccharide nucleoside 6. Further transformations to 9 were accomplished in good
Scheme 1. Disaccharide Building Blocks for Oligonucleotide Synthesis
a) DMTrCl/Py. b) (i-Pr)2NPCl(OCH2CH2CN). c) SnCl4/ClCH2CH2Cl, 0?. d) Bu4NF/THF.
CHEMISTRY & BIODIVERSITY ± Vol. 2 (2005) 1156
overall yields. The structures of the compounds obtained were established by mass
spectrometry and NMR spectroscopy. Most of the chemical shifts and coupling
constants may be calculated directly from NMR spectra. In some cases, comparison
with the published spectra of disaccharide nucleosides , and double-resonance,
1H,13C-correlation and COSY spectra were used for the assignment.
The chemical incorporation of the phosphorylated disaccharide was tested on the T-
loop sequence of yeast initiator tRNAMet (A in Fig. 2). This oligonucleotide
was synthesized with a 3?-propanediol linker. Oligonucleotide assembly using
phosphoramidite 9 was straightforward, provided a suitable coupling time and
activator was used (12 min coupling in the presence of (ethylthio)tetrazole). Selective
removal of the levulinate group was accomplished by exposure to dilute hydrazine for
15 min at room temperature, and allowed, on support, phosphorylation of the attached
ribose moiety. Standard deprotection and purification afforded the modified oligonu-
cleotide, the structure of which was established by LC/ES-MS (mass calc.: 5875.8;
found: 5876.1). The hairpin RNA (A in Fig. 2) has a Tmof 62.7?, while the unmodified
reference hairpin shows a Tm of 65.7? (at 0.1? NaCl). Incorporation of the
phosphorylated disaccharide nucleotide results in a ?Tmof ?3?. In analogy, the
hairpin B depicted in Fig. 2 was synthesized (starting from a rG support), and the
synthesis is outlined in Scheme 2. The quality of the oligonucleotide synthesis is
demonstrated by the HPLC profile of the crude compound after unblocking and
removal from the solid support (Fig. 3). This synthesis shows the compatibility of the
protecting-group chemistry for incorporation of m1A as well as Arp in the same RNA
hairpin. The Tmof the doubly modified RNA hairpin is 61.2?.
It is known that m1A easily undergoes Dimroth rearrangement to yield N6-
methyladenosine. Therefore, we fully analyzed the synthesized oligonucleotide and
established the presence of the intact modified nucleotide by mass spectrometry. The
mass spectrum of the intact oligonucleotides showed that the masses agreed with the
calculated values (calc.: 5671.8; found: 5672.1). The incorporation of m1A was
demonstrated by the fragment-ion spectrum of the base released in the electrospray
CHEMISTRY & BIODIVERSITY ± Vol. 2 (2005) 1157
Fig. 2. Structure of RNA hairpin with Arp (A), and m1A and Arp (B) incorporated
source, as described earlier  (Fig. 4). The ribonuclease-T1digest of the oligonu-
cleotide contained all expected oligonucleotides which also confirms correct assembly
of the sequence. Mass values and retention times are given in the Table.
Conclusions. ± A synthetic strategy was developed for the introduction of 1-
methyladenosine as well as O-?-?-ribofuranosyl-(1???2?)-adenosine-5??-O-phosphate in
RNA. The phosphorylation of the disaccharide nucleoside was carried out after
assembly of the oligonucleotides. Mass spectrometry established the correct incorpo-
ration of both modified nucleosides. This modified oligonucleotide represents the T
loop of the cytoplasmic initiator tRNAMetfrom Schizosaccharomyces pombe . This
T loop contains both a modified nucleotide with a positively charged 1-methyladenine
CHEMISTRY & BIODIVERSITY ± Vol. 2 (2005) 1158
Scheme 2. Synthetic Scheme for Incorporation of Modified Nucleosides in RNA
and a modified nucleotide with a negatively charged phosphate group. The synthesis is
part of a project aimed at elucidating the solution (NMR) structure of the above-
mentioned natural tRNA, including all modified nucleotides. The influence of these
CHEMISTRY & BIODIVERSITY ± Vol. 2 (2005)1159
Fig. 3. HPLC Profile of the crude reaction mixture for the synthesis of the hypermodified RNA B in Fig. 2
Fig. 4. ESI Mass spectrum of the m/z 148 fragment recorded with 20-eV collision energy and cone voltage set at
85 V for base release from 5?-CUCGGAUCGm1AAACCGArpAG-3?. Comparison of this fragment ion spectrum
with spectra obtained from m1A and m6A clearly shows that this oligonucleotide contains m1A.
Table. Mass-Spectrometric Analysis of the Fragments from RNase-T1Digestion of Oligonucleotide B (Fig. 2)
Sequencea) Calculated massb) Found massRetention time [min]
a) Convention: p denotes a 3?-phosphate group, ?p represents a cyclic phosphate.b) Mass calculated using
monoisotopic atom masses.c) The mass value is out of the scan range (m/z 600 to 1500).d) Mass obtained by
deconvolution of the ESI spectrum.
modified nucleotides on the structure of the T loop itself, as well as on the dynamic
interactions between the D loop and the T loop will be investigated. The NMR
structure of the anticodon loop of tRNAMetfrom the same species, incorporating t6A,
has been accomplished recently (publication in preparation). Although the cloverleaf
fold of tRNA is well-studied by X-ray diffraction, little is known about the dynamics of
conformational changes within tRNA, how this is influenced by modified nucleotides,
and how these conformational changes are involved in recognition processes with
synthetases, ribosomes, and other biological components.
General. Column chromatography (CC) was performed on silica gel (0.040±0.063 mm). TLC was carried
out on Kieselgel 260 F (Merck) with detection by UVand the following solvent systems (compositions expressed
as v/v): CH2Cl2(A); CH2Cl2/MeOH 98 :2 (B); CH2Cl2/MeOH 95 :5 (C); detection by UV light. NMR Spectra:
Bruker AMX-400 and Varian Unity-500 NMR spectrometers; at 300 K; chemical shifts ? in ppm rel. to the
solvent signals (1H and13C) or H3PO4as external standard (31P); the coupling constants (J) are given in Hz. The
signals were assigned by double-resonance techniques and COSY experiments. Mass spectrometry and exact
mass measurements of the nucleoside intermediates were performed on a quadrupole/orthogonal-acceleration
time-of-flight tandem mass spectrometer (Q-Tof-2, Micromass, Manchester, UK) equipped with a standard
electrospray ionization (ESI) interface. ESI-MS for the modified oligonucleotides were obtained by coupling
the Q-Tof-2 to a cap. HPLC (CapLC, Waters, Milford, MA). Masses were obtained by deconvolution of the
spectra using the MaxEnt-1 algorithm (MassLynx 3.4, Micromass, Manchester, UK). Chromatographic
conditions: column C18, 0.5 mm?15 mm (PepMap, LC Packings), mobile phase 0.05? Et3N brought to pH 7.5
with 1,1,1,3,3,3-hexafluoropropan-2-ol, with MeCN as the org. phase. Flow rate: 12 ?l min?1. The gradient
started at 2% org. phase and increased at 2% per min over 15 min. For the T1 digest, the gradient started at 0%
and increased at 0.5% over 15 min. Mass spectra were acquired every 2 s in negative-ionization mode applying
2850 V on the electrospray capillary. Cone voltage 35 V, collision cell voltage 10 V. Digestion of the
oligonucleotide (0.55 nmol/?l in 10 m? Tris¥HCl, 1 m? EDTA pH 7.5) was carried out with RNase T1(E.C.
220.127.116.11, USB, Cleveland, USA). 5.5 nmol were hydrolyzed with 11u enzyme for 30 min at 37?. 28 pmol of the
digested oligonucleotide was injected onto the HPLC column. TOM-Amidites and the phosphorylation reagent
were purchased from Glen Research.
yl-?-?-ribofuranosyl}-?-?-ribofuranosyl)guanine (2). Following co-evaporation with anh. pyridine, 630 mg
(0.59 mmol) of the 2?-O-ribosylated guanosine analogue 1 was dissolved in 25 ml of pyridine, and 4,4?-
dimethoxytrityl chloride (241 mg, 0.71 mmol) was added. The mixture was stirred for 3 h at r.t. until TLC
indicated the reaction to be almost completed. Following neutralization with some aq. NaHCO3, the mixture
was concentrated and partitioned twice between CH2Cl2and aq. NaHCO3. The org. layer was purified on 40 g of
silica gel with a MeOH gradient (0 to 2%) in CH2Cl2containing 0.5% of pyridine to afford 680 mg (0.49 mmol,
84%) of 2. Foam.1H-NMR (CDCl3): 12.19 (s, 1 H, NHibu); 10.54 (br. s, N(1)H); 9.11 (s, NH); 8.06 (d, J?8.8,
2 Hmof npe); 8.03 (d, J ? 8.8, 2 Hmof npe); 7.92±7.82 (m, 4 H, Bz); 7.79 (s, H?C(8)); 7.58±7.52 (m, 2 H, Bz);
7.44±7.14 (m, 17 arom. H); 6.78 (d, J?8.8, 2 H, C6H4OMe); 6.76 (d, J?8.8, 2 H, C6H4OMe); 6.31 (d, J(1?,2?)?
2.4, H?C(1?)(Guo)); 5.81 (dd, J(3?,2?)?5.4, J(3?,4?)?6.6, H?C(3?)(Rib)); 5.68 (d, H?C(2?)(Rib)); 5.38
(s, H?C(1?)(Rib)); 4.61 (br. s, H?C(2?), H?C(3?)(Guo)); 4.44 (dt, J(H,P)?6.3, J(H,H)?6.8, CH2OP);
4.38 (dddd, J(4?,5?a)?3.4, J(4?,5?b)?3.9, J(4?,P)?2.0, H?C(4?)(Rib)); 4.29 (ddd, J(5?a,5?b)??11.5, J(5a?,P)?
7.3, Ha?C(5?)(Rib)); 4.22±4.13 (m, H?C(4?)(Guo), CH2OP); 4.06 (ddd, J(5a?,P)?7.3, Hb?C(5?)(Rib));
3.74 (s, 2 MeOC6H4); 3.47 (dd, J(5?a,4?)?2.9, J(5?a,5?b)??10.7, Ha?C(5?)(Guo)); 3.44 (dd, J(5?b,4?)?4.9,
Hb?C(5?)(Guo)); 3.36 (br. s, HO?C(3?)(Guo)); 3.09 (t, J ? 6.6, PhCH2); 2.99 (m, PhCH2); 2.56 (hept., J ? 6.6,
Me2CH); 1.16 (d, Me); 1.10 (d, Me). ESI-MS (pos.): 1374.4291 (C70H69N7O21P?
guanine (3). Compound 2 (640 mg, 0.46 mmol) was dissolved in 6 ml of CH2Cl2under Ar, and EtN(i-Pr)2
(405 ?l, 2.33 mmol) and 2-cyanoethyl N,N-diisopropylchlorophosphoramidite (210 ?l, 0.93 mmol) were added.
The soln. was stirred for 2 h, but the reaction was still incomplete. Further addition of reagents did not improve
1, [M?H]?; calc. 1374.4284).
CHEMISTRY & BIODIVERSITY ± Vol. 2 (2005)1160
the outcome, and H2O (2 ml) was added, the soln. was stirred for 5 min and partitioned between CH2Cl2(50 ml)
and aq. NaHCO3(30 ml). The org. phase was washed with aq. NaCl (2?30 ml) and the aq. phases were back-
extracted with CH2Cl2(20 ml). Evaporation of the org. phase left an oil, which was purified by CC on 40 g of
silica gel (hexane/acetone/Et3N 49 :49 :2) to afford the product as a foam after co-evaporation with CH2Cl2. The
unreacted material was recovered and treated again as described. The obtained fractions were combined,
dissolved in 2 ml of CH2Cl2, and precipitated twice in 60 ml of cold (?70?) hexane to afford 438 mg (0.27 mmol,
60%) of 3. Rf(hexane/acetone/Et3N 39 :59 :2) 0.46.
1,2,3-Tri-O-benzoyl-5-O-levulinyl-?-ribofuranose (4). To a soln. of 1,2,3-tri-O-benzoyl-?-ribofuranose
(; 2 g, 4.33 mmol) in 1,4-dioxane (40 ml), levulinic acid (0.88 ml, 8.66 mmol), DCC (1.78 g, 8.66 mmol), and
DMAP (40 mg) were added. The mixture was kept for 1 h at 20?, and dicyclohexylurea was filtered off and
washed with 1,4-dioxane. The filtrate was evaporated in vacuo, the residue was purified by CC on silica gel to
give 4 (2.21 g, 91%) as an oil, as a 1:1 mixture of anomers. This mixture was separated by CC (100 g) with 1.5%
?-Isomer of 4. Oil. Rf0.83 (B).1H-NMR (CDCl3): 8.13±7.35 (m, 15 H, Bz); 6.65 (s, H?C(1)); 5.92 (d,
J(2,3)?5.0, H?C(2)); 5.87 (dd, J(3,4)?6.9, H?C(3)); 4.73 (ddd, J(4,5a)?4.0, J(4,5b)?5.3, H?C(4)); 4.51
(dd, J(5a,5b)??12.1, Ha?C(5)); 4.31 (dd, Hb?C(5)); 2.63 (m, 2 CH2); 2.11 (s, Me).13C-NMR (CDCl3): 198.44
(MeC?O(Lev)); 172.26 (C?O(Lev)); 165.31, 165.00, 164.58 (C?O(Bz)); 133.75, 133.63, 133.56, 129.94, 129.85,
129.74, 128.59, 128.52, 128.43 (Bz); 98.97 (C(1)); 79.91 (C(4)); 75.03 (C(2)); 71.45 (C(3)); 63.89 (C(5)); 37.74
(CH2); 27.67 (Me).
?-Isomer of 4. Oil. Rf0.71 (B).1H-NMR (CDCl3): 8.08±7.27 (m, 15 H, Bz); 6.93 (d, J(1,2)?4.4, H?C(1));
5.77 (dd, J(3,2)?6.5, J(3,4)?1.9, H?C(3)); 5.66 (dd, H?C(2)); 4.78 (ddd, J(4,5a)?3.1, J(4,5b)?3.7,
H?C(4)); 4.53 (dd, J(5a,5b)??12.1, Ha?C(5)); 4.44 (dd, Hb?C(5)); 2.84 (m, CH2); 2.70 (m, CH2); 2.22 (s,
Me).13C-NMR (CDCl3): 199.10 (MeC?O(Lev)); 172.27 (C?O(Lev)); 165.68, 165.10, 164.97 (C?O(Bz));
133.52, 133.43, 129.92, 129.83, 129.74, 128.43, 128.37 (Bz); 94.90 (C(1)); 82.72 (C(4)); 71.38 (C(2)); 70.77 (C(3));
63.58 (C(5)); 37.92 (CH2); 27.90 (Me).
ribofuranosyl)-?-?-ribofuranosyl]adenine (6). To a cooled soln. (0?) of 4 (918 mg, 1.64 mmol) under N2in 1,2-
dichloroethane (20 ml), SnCl4(0.23 ml, 1.97 mmol) was added, and the soln. was kept at 0? for 10 min. After
addition of N6-benzoyl-9-[3,5-O-(1,1,3,3-tetraisopropyldisiloxane-1,3-diyl)-?-?-ribofuranosyl]adenine (5; 1.01 g,
1.64 mmol), the resulting soln. was kept at 0? for 16 h. The mixture was diluted with CH2Cl2(30 ml), 10% aq.
soln. of NaHCO3(30 ml) was added, and the suspension was stirred at 20? for 20 min. The suspension was
filtered through Hyflo Super Cel, the org. layer was separated, washed with H2O (30 ml), dried (Na2SO4), and
evaporated to dryness. The residue was purified by CC on silica gel (50 g). The column was washed with system
A and elution with 1% MeOH/CH2Cl2gave 6 (1.14 g, 66%). Foam. Rf0.54 (C).1H-NMR (CDCl3): 9.26 (br. s,
NH); 8.77 (s, H?C(8)); 8.34 (s, H?C(2)); 8.05±7.88 (m, 6 H, Bz); 7.60±7.31 (m, 9 H, Bz); 6.19 (s,
H?C(1?)(Ado)); 5.84±5.78 (m, H?C(1?), H?C(2?), H?C(3?)(Rib)); 4.99 (dd, J(3?,2?)?4.7, J(3?,4?)?9.0,
H?C(3?)(Ado)); 4.84 (d, H?C(2?)(Ado)); 4.61 (ddd, J(4?,3?) ? 6.0, J(4?,5?a) ? 3.7, J(4?,5?b) ? 6.9,
H?C(4?)(Rib)); 4.54 (dd, J(5?a,5?b)??11.5, Ha?C(5?)(Rib)); 4.45 (dd, Hb?C(5?)(Rib)); 4.25±4.18 (m,
H?C(4?), Ha?C(5?)(Ado)); 4.05 (dd, J(5?b,4?)?2.3, J(5?b,5?a)??13.2, Hb?C(5?)(Ado)); 2.71±2.66 (m,
CH2(Lev)); 2.56±2.52 (m, CH2(Lev)); 2.13 (s, Me); 1.09±0.97 (m, 28 H, i-Pr).
(MeC?O(Lev)); 172.2 (C?O(Lev)); 165.3, 164.9, 164.6 (C?O(Bz)); 152.7 (C(6)); 150.9 (C(2)); 149.4 (C(4));
142.0 (C(8)); 133.4, 132.6, 129.7, 129.0, 128.8, 128.7, 128.4, 127.8 (Bz); 123.5 (C(5)); 105.59 (C(1?)(Rib)); 88.8
(C(1?)(Ado)); 81.5 (C(4?)(Ado)); 79.6 (C(4?)(Rib)); 78.6 (C(2?)(Ado)); 75.4 (C(2?)(Rib)); 72.4 (C(3?)(Ado));
69.9 (C(3?)(Rib)); 65.2 (C(5?)(Rib)); 59.9 (C(5?)(Ado)); 37.7 (CH2(Lev)); 27.7 (Me(Lev)); 17.4, 17.2, 17.0, 16.8,
16.7, 13.3, 12.7, 12.7, 12.6 (i-Pr).
N6-Benzoyl-9-[2-O-(2,3-di-O-benzoyl-5-O-levulinyl-?-?-ribofuranosyl)-?-?-ribofuranosyl]adenine (7). To
a soln. of 6 (1.14 g, 1.08 mmol) in THF (3 ml), Bu4NF¥3 H2O (950 mg, 3.02 mmol) in THF (3 ml) was added.
After 15 min at 20?, the mixture was concentrated and co-evaporated with CHCl3(2?10 ml). The residue was
applied to CC (SiO2(30 g)). The column was washed with system A and B, and eluted with system C to give 7
(700 mg, 80%). Foam. Rf0.23 (C).1H-NMR (CDCl3): 9.43 (br. s, NH); 8.81 (s, H?C(8)); 8.35 (s, H?C(2));
8.07±7.84 (m, 6 H, Bz); 7.61±7.27 (m, 9 H, Bz); 6.29 (d, J(1?,2?)?7.3, H?C(1?)(Ado)); 6.05 (br. d, J ? 8.0,
HO?C(5?)(Ado)); 5.60 (dd, J(2?,1?)?1.8, J(2?,3?)?5.3, H?C(2?)(Rib)); 5.56 (dd, J(3?,4?)?5.1, H?C(3?)(Rib));
5.23 (dd, J(2?,3?)?4.6, H?C(2?)(Ado)); 5.17 (d, H?C(1?)(Rib)); 4.67 (d, H?C(3?)(Ado)); 4.35 (m, H?C(4?),
Ha?C(5?)(Rib)); 4.31 (s, H?C(4?)(Ado)); 3.96 (d, J(5?a,5?b)? ?12.2, Ha?C(5?)(Ado)); 3.86 (br. s,
HO?C(3?)(Ado)); 3.78 (m, Hb?C(5?)(Rib), Hb?C(5?)(Ado)); 2.78 (m, CH2(Lev)); 2.62 (m, CH2(Lev));
31P-NMR: 151.70, 150.02. ESI-MS (pos.): 1574.5402
2, [M?H]?; calc. 1574.5362).
13C-NMR (CDCl3): 201.8
CHEMISTRY & BIODIVERSITY ± Vol. 2 (2005)1161
2.15 (s, Me).
(C?O(Bz)); 152.00 (C(6)); 150.46 (C(2)); 150.24 (C(4)); 144.51 (C(8)); 133.62, 133.51, 132.73, 129.67, 129.62,
128.75, 128.57, 128.46, 128.39, 128.37, 127.88 (Bz); 124.22 (C(5)); 106.25 (C(1?)(Rib)); 89.23 (C(1?)(Ado)); 87.22
(C(4?)(Ado)); 80.83 (C(2?)(Ado)); 79.92 (C(4?)(Rib)); 75.54 (C(2?)(Rib)); 71.74 (C(3?)(Ado)); 71.52
(C(3?)(Rib??)); 63.52 (C(5?)(Rib)); 63.09 (C(5?)(Ado)); 37.71 (CH2(Lev)); 29.62 (Me(Lev)); 27.68 (CH2(Lev)).
ESI-MS (pos.): 810.2630 (C41H40N5O?
bofuranosyl]adenine (8). Following co-evaporation with anh. pyridine, 7 (596 mg, 0.74 mmol) was dissolved in
pyridine (30 ml), and 4,4?-dimethoxytrityl chloride (287 mg, 0.85 mmol) was added. The mixture was stirred
overnight at r.t. until TLC indicated the reaction to be complete. Following neutralization with 10% aq.
NaHCO3, the mixture was concentrated and partitioned twice between CH2Cl2and aq. NaHCO3. The org. layer
was purified twice on a silica-gel column (40 g) with a MeOH gradient (0 to 1%) in CH2Cl2containing 0.5% of
pyridine, affording a total of 601 mg (0.54 mmol, 73%) of 8. Foam.1H NMR (CDCl3): 9.11 (s, NH); 8.69 (s,
H?C(8)); 8.27 (s, H?C(2)); 8.05±7.89 (m, 6 H, Bz); 7.61±7.14 (m, 18 H, Bz, Ph); 6.78 (d, J ? 8.8, 4 H, Ph); 6.34
(d, J(1?,2?)?5.1, H?C(1?)); 5.65 (dd, J(2?,1?)?1.3, J(2?,3?)?5.2, H?C(2?)(Rib)); 5.60 (dd, J(3?,4?)?6.1,
H?C(3?)(Rib)); 5.36 (d, H?C(1?)(Rib)); 5.31 (dd, J(2?,3?)?4.7, H?C(2?)(Ado)); 4.66 (dd, J(3?,4?)?4.2,
H?C(3?)(Ado)); 4.46 (ddd, J(4?,5?a)?3.7, J(4?,5?b)?4.9, H?C(4?)(Rib)); 4.31 (ddd, J(4?,5?a)?4.0, J(4?,5?b)?
4.7, H?C(4?)(Ado)); 4.25 (dd, J(5?a,5?b)??12.2, Ha?C(5?)(Rib)); 3.97 (dd, Hb?C(5?)(Rib)); 3.76 (s, 2 MeO);
3.52 (dd, J(5?a,5?b)??10.6, Ha?C(5?)(Ado)); 3.44 (dd, Hb?C(5?)(Ado)); 3.12 (d, J ? 4.1, HO?C(3?)(Ado));
2.74±2.71 (m, CH2(Lev)); 2.62±2.59 (m, CH2(Lev)); 2.12 (s, Me).13C NMR (CDCl3; selected signals): 106.5
(C(1?)(Rib)); 87.4 (C(1?)(Ado)); 86.5 (DMTr); 83.9 (C(4?)(Ado)); 80.1 (C(2?)(Ado)); 80.0 (C(4?)(Rib)); 75.8
(C(2?)(Rib)); 71.5 (C(3?)(Ado)); 70.8 (C(3?)(Rib)); 63.5 (C(5?)(Ado)); 63.3 (C(5?)(Rib)); 55.2 (MeO); 37.8
(CH2(Lev)); 29.7 (Me(Lev)); 27.7 (CH2(Lev)). ESI-MS (pos.): 1112.3907 (C62H58N5O?
pyl)amino]phosphinyl}-5-O-dimethoxytrityl-?-?-ribofuranosyl)adenine (9). Compound 8 (600 mg, 0.54 mmol)
was dissolved in 6 ml of CH2Cl2under Ar, and EtN(i-Pr)2(282 ?l, 1.62 mmol) and 2-cyanoethyl N,N-
diisopropylchlorophosphoramidite (268 ?l, 1.2 mmol) were added. After stirring the soln. for 2 h, TLC indicated
the reaction to be complete. Aq. NaHCO3(2 ml) was added, the soln. was stirred for 5 min and partitioned
between CH2Cl2(50 ml) and aq. NaHCO3(30 ml). The org. phase was washed with aq. NaCl soln. (2?30 ml),
and the aq. phases were back-extracted with CH2Cl2(20 ml). Evaporation of the org. phase afforded an oil,
which was purified by CC on 40 g of silica gel (hexane/acetone/Et3N, 66 :32 :2) to afford the product as a foam
after co-evaporation with CH2Cl2. The foam was dissolved in 2 ml of CH2Cl2and precipitated in 100 ml of cold
(?70?) hexane to afford 500 mg (0.38 mmol, 70%) of 9. White powder. Rf(hexane/acetone/Et3N 49 :49 :2) 0.43.
31P-NMR: 150.94, 150.81. ESI-MS (pos.): 1312.5027 (C71H75N7O16P?
Oligoribonucleotide Synthesis. Oligonucleotide synthesis was performed on an Expedite DNA synthesizer
(Applied Biosystems) by the phosphoramidite approach. The standard 1-?mol scale RNA assembly protocol
was used in combination with TOM¾amidites and 0.25? (ethylthio)tetrazole (ETT) for activation (13-min
coupling time). The glycosylated analogue was used at 0.07? with a total coupling time of 15 min. The
oligonucleotide was fully assembled leaving the DMTr moiety on, and subsequently the support was treated
manually via syringe with a NH2NH2soln. (40 ?l of NH2NH2¥H2O, 2 ml of pyridine, and 0.5 ml of dry AcOH)
for 15 min at r.t. The support was placed in line on the synthesizer again, washed thoroughly with MeCN, and
coupling was carried out with a 0.12? soln. of the sulfonyl phosphorylating agent (Glen Research 10-1900-02)
and 0.25? of ETT (10 min). Standard detritylation removed both DMTr moieties. The oligonucleotide was
cleaved from the support and deprotected as recommended for TOM-protected nucleotides. Hereto, the
support was treated with 0.75 ml each of a 40% aq. MeNH2soln. and a 8? MeNH2soln. in EtOH for 6 h at 35?.
Following lyophilization of the obtained soln., the residue was treated with 1 ml of a 1? Bu4NF soln. in THF, by
heating for 10 min at 55?, followed by 24 h at r.t. The soln. was neutralized with 1 ml of 1? Tris buffer and
concentrated to half volume again, after which gel filtration was carried out on a NAP-25¾column (Sephadex
G25-DNA-grade, Pharmacia). Purification was achieved on a Mono-Q¾HR 10/10 anion-exchange column
(Pharmacia) with the following gradient system: A: 10 m? NaClO4and B: 600 m? NaClO4, both in 20 m? Tris¥
HCl, 15% MeCN, 0.1 m? EDTA at pH 7.1; flow rate 2 ml min?1. The product-containing fraction was desalted
on a NAP-25¾column and lyophilized. Conditions for determining Tmwere described in .
13C-NMR (CDCl3): 207.05 (MeC?O(Lev)); 172.35 (C?O(Lev)); 165.28, 165.25, 164.61
13, [M?H]?; calc. 810.2622).
15, [M?H]?; calc.
9, [M?H]?; calc. 1312.5008).
The authors thank the KUL Research Council (GOA), NATO, Russian Foundation for Basic Research, and
RAS Programme −Molecular and Cellular Biology× for financial support.
CHEMISTRY & BIODIVERSITY ± Vol. 2 (2005) 1162
 J. Rozenski, P. F. Crain, J. A. McCloskey, Nucleic Acids Res. 1999, 27, 196.
 P. F. Agris, Nucleic Acids Res. 2004, 32, 223.
 P. F. Agris, Prog. Nucleic Acid Res. 1996, 53, 79.
 Y. A. Wang, Acc. Cem. Res. 1984, 17, 393.
 K. K. Ogilvie, N. Usman, K. Nicoghosian, R. J. Cedergren Proc. Natl. Acad. Sci. U.S.A. 1988, 85, 5764.
 D. Gasparutto, T. Livache, H. Bazin, A.-M. Duplaa, A. Guv, A. Khorlin, D. Molko, A. Roget, R. Teoule
Nucleic Acids Res. 1992, 20, 5159.
 T. E. England, O. C. Uhlenbeck, Biochemistry 1978, 17, 2069.
 T. Ohtsuki, G. Kawai, Y. Watanabe, K. Kita, K. Nishikawa, K. Watanabe, Nucleic Acids Res. 1996, 24, 662.
 J. A. Preiss, A. D. Wolfson, O. C. Uhlenbeck, Biochemistry 2000, 39, 8250.
 G. Keith, A.-L. Glasser, J. Desgres, K. C. Kuo, C. W. Gehrke, Nucleic Acids Res. 1990, 18, 5989.
 S. Kiesewetter, G. Ott, M. Sprinzl, Nucleic Acids Res. 1990, 18, 4677.
 A.-L.Glasser, J. Desgres, J. Heitzler, C. W. Gehrke, G. Keith, Nucleic Acids Res. 1991, 19, 5199.
 G. Keith, J. Heitzler, C. E. Adlouni, A.-L. Glasser, C. Fix, J. Desgres, G. Dirheimer, Nucleic Acids Res.
1993, 21, 2949.
 T. Basavappa, P. B. Sigler, EMBO J. 1991, 10, 3105.
 H. Sierzputowska-Gracz, H. D. Gopal, P. F. Agris, Nucleic Acids Res. 1986, 14, 7783.
 M. Helm, H. Brule, F. Degoul, C. Cepanec, J. P. Leroux, R. Giege, C. Florentz, Nucleic Acids Res. 1998, 26,
 M. Helm, R. Giege ¬, C. Florentz, Biochemistry 1999, 38, 13338.
 S. N. Mikhailov, J. Rozenski, E. V. Efimtseva, R. Busson, A. Van Aerschot, P. Herdewijn, Nucleic Acids
Res. 2002, 30, 1124.
 E. V. Efimtseva, S N. Mikhailov, Rev. Biochem. (Moscow) 2002, 67, 1136.
 E. V. Efimtseva, S. N. Mikhailov, Russ. Chem. Rev. 2004, 73, 401.
 K. Nauwelaerts, E. V. Efimtseva, S. N. Mikhailov, P. Herdewijn, in −Frontiers in Nucleosides and Nucleic
Acids×, Eds. R. F. Schinazi, D. C. Liotta, IHL Press, 2004, p. 187±220.
 W. T. Markiewicz, A. Niewczyk, Z. Gdaniec, D. A. Adamiak, Z. Dauter, W. Rypniewski, M. Chmielewski,
Nucleosides Nucleotides 1998, 17, 411.
 I. Luyten, R. M. Esnouf, S. N. Mikhailov, E. V. Efimtseva, P. Michiels, H. A. Heus, C. W. Hilbers, P.
Herdewijn, Helv. Chim. Acta 2000, 83, 1278.
 A. A. Rodionov, E. V. Efimtseva, S. N. Mikhailov, J. Rozenski, I. Luyten, P. Herdewijn, Nucleosides
Nucleotides Nucleic Acids 2000, 19, 1847.
 E. V. Efimtseva, A. A. Shelkunova, S. N. Mikhailov, K. Nauwelaerts, J. Rozenski, E. Lescrinier, P.
Herdewijn, Helv. Chim. Acta 2003, 86, 504.
 F. Himmelsbach, R. Charubala, W. Pfleiderer, Helv. Chim. Acta 1987, 70, 1286.
 I. V. Gulyaeva, E. V. Efimtseva, A. A. Rodionov, B. S. Ermolinsky, S. N. Mikhailov, Collect. Czech. Chem.
Commun. Symp. Ser. 2002, 5, 312.
 P. J. L. M. Quaedflieg, C. M. Timmers, V. E. Kal, G. A. van der Marel, E. Kuyl-Yeheskiely, J. H. van Boom,
Tetrahedron Lett. 1992, 33, 3081.
 M. Grotli, R. Eritja, B. Sproat, Tetrahedron 1997, 53, 11317.
 M. H. Lyttle, T. A. Walton, D. J. Dick, T. G. Carter, J. H. Beckman, R. M. Cook, Bioconjugate Chem. 2002,
 L. N. Beigelman, S. N. Mikhailov, Carbohydr. Res. 1990, 203, 324.
 S. N. Mikhailov, E. V. Efimtseva, G. V. Gurskaya, V. E. Zavodnik, A. De Bruyn, J. Rozenski, P. Herdewijn,
J. Carbohydr. Chem. 1997, 16, 75.
 M. Simsek, U. L. RajBhandary, Biochem. Biophys. Res. Commun. 1972, 49, 508.
Received May 26, 2005
CHEMISTRY & BIODIVERSITY ± Vol. 2 (2005) 1163