O6-(benzotriazol-1-yl)inosine derivatives: Easily synthesized, reactive nucleosides

Article abstract of DOI:10.1021/ja064682n

A novel class of O⁶-(benzotriazol-1-yl)inosine as well as the corresponding 2′-deoxy derivatives can be conveniently prepared by a reaction between sugar-protected or -unprotected inosine or 2′- deoxyinosine nucleosides and 1H-benzotriazol-1-yloxy-tris(dimethylamino) phosphonium hexafluorophosphate (BOP). The reaction appears to proceed via a nucleoside phosphonium salt, and in the absence of any additional nucleophile, the released 1-hydroxybenzotriazole undergoes reaction with the formed phosphonium salt leading to the requisite O⁶-(benzotriazol-1-yl) inosine or 2′-deoxyinosine derivatives. Isolation and characterization of the phosphonium salt as well as analysis by ³¹P{¹H} NMR appear to be consistent with this reaction pathway. The resulting O ⁶-(benzotriazol-1-yl)inosine derivatives are effective as electrophilic nucleosides, undergoing facile reactions with a variety of nucleophiles such as alcohols, phenols, amines, and a thiol. Unusual and challenging nucleoside derivatives such as an aryl-bridged dimer, a nucleoside-amino acid conjugate, and a nucleoside- nucleoside dimer have also been synthesized from the O⁶-(benzotriazol-1-yl)-2′- deoxyinosine derivative. Finally, a fully protected DNA building block, the O⁶-(benzotriazol-1-yl)-2′-deoxyinosine 5′-O-DMT 3′-O-phosphoramidite, has been prepared and a preliminary evaluation of its use for DNA modification has been performed. Results from these studies indicate several important facts: A single, simple methodological approach provides a class of stable, isolable ribo and 2′-deoxyribonucleoside derivatives that possess excellent reactivity for S_NAr chemistry with a wide range of nucleophiles. Also, a benzotriazolyl nucleoside phosphoramidite appears to be a suitable reagent for incorporation into DNA for purposes of site-specific DNA modification.

Full text of DOI:10.1021/ja064682n

Published on Web 01/09/2007
O6-(Benzotriazol-1-yl)inosine Derivatives: Easily Synthesized,
Reactive Nucleosides
Suyeal Bae and Mahesh K. Lakshman*
Contribution from the Department of Chemistry, The City College and The City UniVersity of
New York, 138th Street at ConVent AVenue, New York, New York 10031-9198
Received July 1, 2006; E-mail: lakshman@sci.ccny.cuny.edu
Abstract: A novel class of O6-(benzotriazol-1-yl)inosine as well as the corresponding 2′-deoxy derivatives
can be conveniently prepared by a reaction between sugar-protected or -unprotected inosine or
2′-deoxyinosine nucleosides and 1H-benzotriazol-1-yloxy-tris(dimethylamino)phosphonium hexafluoro-
phosphate (BOP). The reaction appears to proceed via a nucleoside phosphonium salt, and in the absence
of any additional nucleophile, the released 1-hydroxybenzotriazole undergoes reaction with the formed
phosphonium salt leading to the requisite O6-(benzotriazol-1-yl)inosine or 2′-deoxyinosine derivatives.
Isolation and characterization of the phosphonium salt as well as analysis by 31P{1H} NMR appear to be
consistent with this reaction pathway. The resulting O6-(benzotriazol-1-yl)inosine derivatives are effective
as electrophilic nucleosides, undergoing facile reactions with a variety of nucleophiles such as alcohols,
phenols, amines, and a thiol. Unusual and challenging nucleoside derivatives such as an aryl-bridged dimer,
a nucleoside-amino acid conjugate, and a nucleoside-nucleoside dimer have also been synthesized from
the O6-(benzotriazol-1-yl)-2′-deoxyinosine derivative. Finally, a fully protected DNA building block, the O6-
(benzotriazol-1-yl)-2′-deoxyinosine 5′-O-DMT 3′-O-phosphoramidite, has been prepared and a preliminary
evaluation of its use for DNA modification has been performed. Results from these studies indicate several
important facts: A single, simple methodological approach provides a class of stable, isolable ribo and
2′-deoxyribonucleoside derivatives that possess excellent reactivity for SNAr chemistry with a wide range
of nucleophiles. Also, a benzotriazolyl nucleoside phosphoramidite appears to be a suitable reagent for
incorporation into DNA for purposes of site-specific DNA modification.
Scheme 1. Nucleoside Modification by Nucleophilic Displacement
Introduction
Convenient access to modified nucleosides continues to be
of high importance due to the wide-ranging pharmacological,
biochemical, and biological importance of these compounds.
For nucleoside modification, electrophilic nucleoside derivatives
are an important class of compounds because they serve as
precursors to other nucleoside analogues via nucleophilic
displacement (SNAr) reactions (Scheme 1).1
For this reason, substantial research effort has been directed
toward the synthesis of stable but relatively reactive nucleoside
derivatives bearing leaving groups. For modification at the C-6
position of purine nucleosides, the most common electrophilic
precursors are the halo derivatives,2-5 but phenoxy,6 aryl, and
alkyl sulfonyl,6c,7 pyridyl,8 sulfone,9 and imidazolyl9,10 deriva-
tives have all found applications. Syntheses of many of these
compounds are nontrivial and can pose significant preparative
challenges. Because of these reasons, recently there were two
separate reports on the utility of nucleoside phosphonium salts
as reactive, electrophilic derivatives. In one methodology, the
C-6 amide carbonyl of inosine nucleosides was converted to
the putative triphenylphosphonium salt 2 via reaction with PPh3/
(2) Examples of chloro nucleoside synthesis: (a) Robins, M. J.; Basom, G. L.
Can. J. Chem. 1973, 51, 3161-3169. (b) Robins, M. J.; Uznan´ski, B. Can.
J. Chem. 1981, 59, 2601-2607. (c) Ve´liz, E. A.; Beal, P. A. Tetrahedron
Lett. 2000, 41, 1695-1697. (d) Francom, P.; Janeba, Z.; Shibuya, S.;
Robins, M. J. J. Org. Chem. 2002, 67, 6788-6796. (e) Francom, P.; Robins,
M. J. J. Org. Chem. 2003, 68, 666-669.
(3) Examples of bromo nucleoside synthesis: (a) Nair, V.; Richardson, S. G.
J. Org. Chem. 1980, 45, 3969-3974. (b) Lakshman, M. K.; Keeler, J. C.;
Hilmer, J. H.; Martin, J. Q. J. Am. Chem. Soc. 1999, 121, 6090-6091. (c)
Ve´liz, E. A.; Beal, P. A. J. Org. Chem. 2001, 66, 8592-8598. (d) Lagisetty,
P.; Russon, L. M.; Lakshman, M. K. Angew. Chem., Int. Ed. 2006, 45,
3660-3663.
(4) Examples of iodo nucleoside synthesis: (a) Cosstick, R.; Douglas, M. E.
J. Chem. Soc., Perkin Trans. 1 1991, 1035-1040. (b) van der Wenden,
E. M.; von Frijtag Drabbe Ku¨nzel, J. K.; Mathoˆt, R. A. A.; Danhof, M.;
IJzerman, A. P.; Soudijn, W. J. Med. Chem. 1995, 38, 4000-4006. (c)
Liu, J.; Janeba, Z.; Robins, M. J. Org. Lett. 2004, 6, 2917-2919.
(5) Examples of fluoro nucleoside synthesis: ref 2a. Robins, M. J.; Uznan´ski,
B. Can. J. Chem. 1981, 59, 2608-2611.
(1) For some representative examples on the use of SNAr chemistry over the
years for nucleoside modification, please see: (a) Nucleosides and
Nucleotides; Townsend, L. B., Ed.; Plenum Press: New York, 1988; Vol.
1, Chapter 2, pp. 113-281. (b) Gerster, J. F.; Robins, R. K. J. Am. Chem.
Soc. 1965, 87, 3752-3759. (c) Lakshman, M.; Lehr, R. E. Tetrahedron
Lett. 1990, 31, 1547-1550. (d) Samano, V.; Miles, R. W.; Robins, M. J.
J. Am. Chem. Soc. 1994, 116, 9331-9332. (e) Kim, S. J.; Jajoo, H. K.;
Kim, H.-Y.; Zhou, L.; Horton, P.; Harris, C. M.; Harris, T. M. Bioorg.
Med. Chem. 1995, 3, 811-822. (f) Chaturvedi, S.; Lakshman, M. K.
Carcinogenesis 1996, 17, 2747-2752. (g) Choi, Y.; George, C.; Strazewski,
P.; Marquez, V. E. Org. Lett. 2002, 4, 589-592. (h) Jeong, L. S.; Jin,
D. Z.; Kim, H. O.; Shin, D. H.; Moon, H. R.; Gunaga, P.; Chun, M. W.;
Kim, Y.-C.; Melman, N.; Gao, Z.-G.; Jacobson, K. A. J. Med. Chem. 2003,
46, 3775-3777.
9
782
J. AM. CHEM. SOC. 2007, 129, 782-789
10.1021/ja064682n CCC: $37.00 © 2007 American Chemical Society

O6-(Benzotriazol-1-yl)inosine Derivatives
A R T I C L E S
Scheme 2. In situ Formation of Nucleoside Phosphonium Salts as
Electrophilic Nucleosides
Scheme 3. Two Possible Reaction Modes of Inosine Nucleosides
with BOP
I2/(iso-Pr)2NEt that underwent displacement chemistry in the
presence of an amine, to yield adenine derivatives.9 In the second
method, reaction of the C-6 amide carbonyl with 1H-benzo-
triazol-1-yloxy-tris(dimethylamino)phosphonium hexafluoro-
phosphate (BOP) and (iso-Pr)2NEt led to the postulated forma-
tion of a tris(dimethylamino)phosphonium salt 3 that could be
converted to adenine derivatives 4 with an amine.11 Both
methods were one-pot procedures that do not allow for the
isolation of the phosphonium salt. Scheme 2 shows these two
parallel approaches. In this context, a serendipitous discovery
was reported wherein guanine residues undergo reaction with
PyBOP at the C-6 position during PNA assembly, resulting in
a similar phosphonium salt that then undergoes substitution by
piperidine during Fmoc cleavage.12
It was at this point that we became interested in the reactions
of inosine nucleosides with BOP, and during the course of these
investigations, we have discovered the novel class of O6-
(benzotriazol-1-yl)inosine derivatives that are effective, elec-
trophilic nucleosides. For many applications, these new stable,
isolable derivatives could replace other electrophilic nucleosides
that are more difficult to access. Furthermore, these compounds
can potentially be used in reactions with a wide range of
nucleophiles that are perhaps not possible with methodology
involving the in situ generation of phosphonium salts. This
report describes the synthesis of these novel benzotriazolyl
inosine and 2′-deoxyinosine derivatives, a plausible mechanism
of formation, their reactions with a range of nucleophiles, and
the synthesis of a suitable building block for DNA assembly
and its potential use for site-specific DNA modification.
to adenine derivatives in the presence of an amine, we
questioned whether other nucleophiles could be utilized in this
reaction. Oxygen nucleophiles were interesting, but competing
reactions at the phosphorus atom in the formed phosphonium
salt were something that elicited our concern. In conjunction
with the possibility of introducing an oxygen nucleophile, we
also wondered if a moiety could be introduced that would serve
as a leaving group in subsequent SNAr displacement reactions.
It has been reported that 1-hydroxybenzotriazole (HOBT)
derived adducts were observed by LC/MS in the reactions of
inosine with BOP and aryl amines.11,13 This led us initially to
question the mechanism of this reaction.
As shown in the example in Scheme 3, there are two possible
modes by which the hypoxanthine core of inosine nucleosides
could react with BOP. In pathway (a), the reaction could be
initiated via attack of the amide oxygen of the nucleoside at
the phosphorus atom of BOP, and alternatively in pathway (b),
the reaction could occur between the amide oxygen and the ring
nitrogen of BOP (SN2′-like). In pathway (a), the intermediate
would indeed be the hexamethylphosphonium salt 3a, whereas
in pathway (b), the O6-(benzotriazol-1-yl) product 5a would be
expected to result from expulsion of HMPA in a single step. In
pathway (a), the alkoxide formed from HOBT could then
displace HMPA from the hexamethylphosphonium salt 3a, also
leading to the O6-benzotriazol-1-yl product 5a.
To make an assessment, two courses of action were consid-
ered: (i) ascertain whether the hexamethylphosphonium salt
could be observed under short reaction periods and (ii) evaluate
the course of the reaction by 31P{1H} NMR to look for
intermediates and products. First, 3′,5′-bis-O-(tert-butyldimeth-
ylsilyl)-2′-deoxyinosine (1a) was allowed to react with 2 molar
equiv of BOP and 1.5 molar equiv of (iso-Pr)2NEt in dry THF
at room temperature for 4 h. Workup and purification of the
products by preparative TLC led to the isolation of phosphonium
salt 3a, albeit in low yield. Compound 3a displayed a doublet
Results and Discussion
Because it had been documented in the literature that PPh3/
I2 and BOP underwent reaction with inosine nucleosides leading
(6) (a) O6-(Pentafluorophenyl)-2′-deoxyguanosine: Gao, H.; Fathi, R.; Gaffney,
B. L.; Goswami, B.; Kung, P.-P.; Rhee, Y.; Jin, R.; Jones, R. A. J. Org.
Chem. 1992, 57, 6954-6959. (b) O6-(4-Bromophenyl)-2′-deoxyinosine and
O6-(phenyl)-2′-deoxyinosine: Lakshman, M. K.; Zajc, B. Nucleosides
Nucleotides 1996, 15, 1029-1039. (c) O6-(4-Chlorophenyl)inosine: Al-
lerson, C. R.; Chen, S. L.; Verdine, G. L. J. Am. Chem. Soc. 1997, 119,
7423-7433.
for the N(CH3)2 resonances centered at δ 2.92 ppm (JP-H
)
(7) (a) Hayakawa, Y.; Hirose, M.; Noyori, R. J. Org. Chem. 1993, 58, 5551-
5555. (b) Nagatsugi, F.; Uemura, K.; Nakashima, S.; Maeda, M.; Sasaki,
S. Tetrahedron 1997, 53, 3035-3044.
11.2 Hz). In comparison, the N(CH3)2 resonance of BOP appears
at δ 2.86 ppm (JP-H ) 10.3 Hz).14 This confirmed the presence
(8) (a) Adamiak, R. W.; Biała, E.; Skalski, B. Nucleic Acids Res. 1985, 13,
2989-3003. (b) Fathi, R.; Goswami, B.; Kung, P.-P.; Gaffney, B. L.; Jones,
R. A. Tetrahedron Lett. 1990, 31, 319-322.
(13) While this work was in progress, it was reported that reaction of
4-hydroxyquinazoline with BOP and DBU in CH3CN led to the formation
of 4-(benzotriazol-1-yloxy)quinazoline: Wan, Z.-K.; Wacharasindhu, S.;
Binnun, E.; Mansour, T. Org. Lett. 2006, 8, 2425-2428. In reactions with
the less nucleophilic aryl amine, HOBT adducts were also observed in this
report.
(9) Lin, X.; Robins, M. J. Org. Lett. 2000, 2, 3497-3499.
(10) Janeba, Z.; Lin, X.; Robins, M. J. Nucleosides Nucleotides Nucleic Acids
2004, 23, 137-147.
(11) Wan, Z.-K.; Binnun, E.; Wilson, D. P.; Lee, J. Org. Lett. 2005, 7, 5877-
5880.
(12) Pritz, S.; Wolf, Y.; Klemm, C.; Bienert, M. Tetrahedron Lett. 2006, 47,
5893-5896.
(14) The N(CH3)2 resonance of HMPA appears at δ 2.63 ppm (JP-H ) 9.5 Hz),
and that of HMPT appears at δ 2.47 ppm (JP-H ) 9.2 Hz).
9
J. AM. CHEM. SOC. VOL. 129, NO. 4, 2007 783

A R T I C L E S
Bae and Lakshman
Figure 1. 31P{1H} NMR studies. Panel A: NMR spectra of BOP, phosphonium salt 3a, and HMPA. Panel B (spectra offset to the right): time course of
the reaction between 1a, BOP, and (iso-Pr)2NEt.
Scheme 4. Synthesis of O6-(Benzotriazol-1-yl)inosine Derivatives
of the phosphonium salt (3a in Scheme 3) in the reaction mixture
and indicated that pathway (a) was likely in operation.
Finally, at 24 h and then at 60 h, the resonance from 3a
decreases and the HMPA resonance is prominent.
To conduct an analysis of the reaction by 31P{1H} NMR, the
31P chemical shifts of BOP, nucleoside phosphonium salt 3a,
and HMPA (Figure 1, panel A) were unambiguously ascertained.
Next, the 31P{1H} NMR spectra of the reaction mixture were
recorded as a function of time (Figure 1, panel B). For this, a
reaction among 1a, 2 molar equiv of BOP, and 1.5 molar equiv
of (iso-Pr)2NEt was conducted in CDCl3 at room temperature
in an NMR tube. As can be seen from the stacked NMR plots,
at 20 min there is a clear discernible resonance corresponding
to salt 3a at δ 35.44 ppm, but the extent of formation of HMPA
remains very minor. At 1 h, as the resonance from 3a becomes
more intense, the HMPA resonance continues to remain minor.
These results clearly indicated that HMPA formation is not
a rapid occurrence and that there is intermediacy of the
phosphonium salt 3a. Notably, even at 60 h in CDCl3, a small
trace of 3a persists. With these data in hand, the next step was
the determination of structure of the product in the reaction of
BOP with 1a (Scheme 4). Prolonged reaction (23 h) of 1a with
BOP and (iso-Pr)2NEt in THF yielded a single compound in
1
83% isolated yield. The H and 13C NMR data of this product
were consistent with the O6-(benzotriazol-1-yl) derivative 5a.
Unambiguous structural characterization of 5a came via the
reaction of bromo nucleoside 6 with HOBT under conditions
we have previously utilized for the synthesis of O6-alkyl and
aryl 2′-deoxyguanosine derivatives.15 The product derived from
the reaction of 6 with HOBT was identical to that derived from
(15) Lakshman, M. K.; Ngassa, F. N.; Keeler, J. C.; Dinh, Y. Q. V.; Hilmer,
J. H.; Russon, L. M. Org. Lett. 2000, 2, 927-930.
9
784 J. AM. CHEM. SOC. VOL. 129, NO. 4, 2007

O6-(Benzotriazol-1-yl)inosine Derivatives
A R T I C L E S
the reaction of 1a with BOP. With a mechanistic understanding
and structural proof completed, the next stage was the synthesis
of other related benzotriazolyl nucleoside derivatives. Using the
optimized conditions, we converted fully protected inosine to
the O6-(benzotriazol-1-yl) derivative 5b in 80% yield. Because
direct BOP-mediated aminations of unprotected nucleosides
have been reported (seven examples with inosine and one with
2′-deoxyinosine),11 it was reasoned that the hydroxyl unprotected
O6-benzotriazolyl derivatives should also be available via this
procedure. Correspondingly, both 5c and 5d could be obtained
in 57% and 53% yields, respectively (it is possible that some
losses could have occurred due to water solubility). However,
for solubility considerations, these two reactions were performed
in DMF rather than in THF. Generally, these transformations
appear to be quite scalable and have been performed on the
0.5-3 g scales.
With the syntheses of O6-(benzotriazol-1-yl)inosine deriva-
tives completed, the next stage was an assessment of their use
as substrates for SNAr displacement reactions. Displacement of
HOBT from the C-6 position of a purine with amines has been
reported in the preparation of a HOBT linker for heterocycle
synthesis.16 For the present study a range of alcohols, phenols,
amines, and a thiol were selected for displacement reactions.
For the synthesis of simple alkyl ethers, alcohols were chosen
as the reaction medium, whereas with other nucleophiles, toluene
or DME was used. The reaction temperature ranged from room
temperature to 105 °C depending upon the nucleophile; Cs2-
CO3 was used as the base for all reactions except in the case of
p-toludine as well as some reactions with morpholine and
benzylamine. The results from these experiments are compiled
in Table 1.
Simple nucleophiles underwent smooth coupling with nucleo-
side derivatives 5a-d, and some observations are noteworthy.
Reaction of 5a with MeOH does not proceed at room temper-
ature (18 h) or at 85 °C (24 h) in the absence of a base, clearly
indicating that these substrates are stable to solvolysis. On the
other hand, as can be anticipated, 5a and 5b undergo reaction
with stronger nucleophiles such as morpholine and benzyl amine
in the absence of base. With p-toluidine (entries 23-26), after
testing several conditions, the use of EtOH/(iso-Pr)2NEt proved
optimal. Solvents such as DMF and DME led to little desirable
product, and the combination of EtOH/Cs2CO3 could not be
used. In the previously reported one-pot reaction of inosine with
BOP and aryl amines, it has been noted, albeit in the Supporting
Information, that the reactions were conducted in DMF followed
by dilution with EtOH and heating.11 Although this procedural
aspect was initially not clear to us, it is now obvious that EtOH
is indeed a suitable solvent for the displacement step of this
reaction, and as we have demonstrated with MeOH, simple
alcoholysis of the benzotriazolyl nucleoside does not occur.
Also, notably, it has been reported that a trissilyl-protected
6-bromopurine nucleoside was unreactive toward p-toluidine and
that the 6-chloro analogue was poorer reacting in SNAr
displacements compared to the bromo nucleoside.3c Even with
morpholine, the trissilyl 6-bromopurine nucleoside was slow
to react and underwent reaction only in 9 h in DME.3c In contrast
to these results, facile reactions were attained with both 5a and
5b in this study, perhaps indicating the superiority of these
benzotriazolyl nucleosides over the corresponding bromo and
chloro analogues as substrates for displacement reactions. The
unprotected nucleosides were also tested for their reactivity
(entries 16, 17, 25, and 26), and although the yields are lower,
possibly due to high polarity of the products, successful reactions
were attained.
The reactivity of 5a was then challenged via its use in unusual
reactions. As shown in Scheme 5, we queried the reactions of
5a with a bisphenol (hydroquinone 11), an amino acid [(S)-N-
Boc-tyrosine methyl ester 12], and a nucleoside (3′,5′-bis-O-
(tert-butyldimethylsilyl)-2′-deoxyinosine 1a).
With 2.5 molar equiv each of 5a and Cs2CO3, the reaction
of hydroquinone (11) was complete within 3 h in DME at 85
°C, affording the aryl-bridged dimer 13 in 68% yield. In the
reaction with (S)-N-Boc-tyrosine methyl ester (12), epimeriza-
tion is possible under the basic reaction conditions. Therefore,
the optical rotation of 12 was initially determined under various
conditions (Table 2). These experiments showed that partial
epimerization did occur under the reaction conditions with a
greater loss of chirality occurring with Cs2CO3. However, no
loss of chirality was experienced with NaHCO3.
Having ascertained this, reactions of 5a with 12 were
conducted. In DME with Cs2CO3 at 85 °C, a 64% yield of 14
was obtained over 1 h whereas use of NaHCO3 in place of Cs2-
CO3 led to a longer (24 h) reaction and a lowered 41% yield.
1
Although the H NMR spectrum of 14 resulting from the Cs2-
CO3-mediated reaction did not show distinct signals correspond-
ing to two diastereomers, results in Table 2 indicate possible
epimerization. On the other hand, 14 arising from the NaHCO3-
mediated reaction is most likely a single diastereomer. The point
of this experiment was to demonstrate that unusual peptide-
nucleoside conjugates can also be prepared via the method.
Another observation that emerged from the reactions of 5a with
12 was that base was needed for this reaction with a phenolic
hydroxyl group as well, and no reaction was observed in the
absence of a base over 17 h at 85 °C. Finally, the reaction of
5a with 1a was conducted in the presence of Cs2CO3 in DME
at 85 °C, resulting in a 60% yield of the unsymmetrical dimer
15. Use of K3PO4 in this reaction proved to be better, providing
15 in 92% yield. The structure of the unsymmetrical dimer was
obvious from its 1H NMR spectrum, where dual resonances were
observed. A symmetrical dimeric structure at the O atom would
have produced only a single set of resonances. Interestingly,
5a and 5b failed to produce any product in reactions with 3′,5′-
bis-O-(tert-butyldimethylsilyl)-2′-deoxyadenosine.
Once the displacement reactions were completed, synthesis
of a suitable DNA building block bearing the O6-(benzotriazol-
1-yl) unit was assessed. For this, two options were available:
(a) conversion of 5′-O-DMT-2′-deoxyinosine to the BOP
derivative or (b) conversion of O6-(benzotriazol-1-yl)-2′-deoxyi-
nosine (5c) to the 5′-O-DMT derivative, followed in each case
by conversion to the 3′-O-phosphoramidite. Alternative (a) was
selected because this would also allow a test of the compatibility
of the DMT protecting group with the reaction conditions
leading to the benzotriazolyl nucleoside derivative. As shown
in Scheme 6, reaction of the known 5′-O-DMT-2′-deoxyinosine
1617 with 2 mol equiv each of BOP and (iso-Pr)2NEt at room
temperature over 40 h led to the O6-(benzotriazol-1-yl)-5′-O-
DMT-2′-deoxyinosine 17 in 85% isolated yield. Finally, conver-
(16) Scicinski, J. J.; Congreve, M. S.; Jamieson, C.; Ley, S. V.; Newman,
E. S.; Vinader, V. M.; Carr, R. A. E. J. Comb. Chem. 2001, 3, 387-396.
(17) Seela, F.; Kaiser, K. Nucleic Acids Res. 1986, 14, 1825-1844.
9
J. AM. CHEM. SOC. VOL. 129, NO. 4, 2007 785

A R T I C L E S
Bae and Lakshman
Table 1. Reactions of Nucleophiles with O6-(Benzotriazol-1-yl)inosine Nucleosides 5a-da
a See the Supporting Information for details on specific reactions.
9
786 J. AM. CHEM. SOC. VOL. 129, NO. 4, 2007

O6-(Benzotriazol-1-yl)inosine Derivatives
A R T I C L E S
Scheme 5. Reactions of 5a with Hydroquinone,
(S)-N-Boc-tyrosine Methyl Ester, and 2′-Deoxyinosine
suitability of phosphoramidite 18 for DNA incorporation and
for displacement reactions post-assembly. Additional detailed
studies on the use of 18 for DNA synthesis and post-assembly
modification of DNA, as well as other applications of these
modified nucleosides, are currently being considered.
In summary, we have demonstrated that a family of novel
O6-(benzotriazol-1-yl)inosine nucleosides can be synthesized in
a facile and simple manner, where a unified method provides
both ribo and deoxyribo derivatives. These stable, storable
compounds possess excellent reactivity toward a range of
nucleophiles, perhaps superior to the corresponding bromo and
chloro derivatives, making them well suited for nucleoside and
DNA modification. The ease of preparation of these compounds
is also noteworthy from several other angles. Protection-
deprotection strategies for the saccharide can be circumvented.
When compared to the synthesis of reactive O6-sulfonyl
derivatives of inosine nucleosides, where competing sulfony-
lation at N1 and O6 is observed,6c,18 reactions leading to the
benzotriazolyl nucleoside derivatives occur exclusively at the
O6. This particular feature avoids loss of a valuable precursor.
Finally, the method is compatible with the DMT protecting
group and is well suited for the synthesis of a DNA building
block that can in turn be used for DNA assembly and post-
assembly DNA modification. It is anticipated that the chemistry
described will lead to the commercial availability of these
modified nucleosides.19
Table 2. Exposure of (S)-N-Boc-tyrosine Methyl Ester 12 to
Various Conditionsa
Experimental Section
entry
solvent
base
temp
time
[R]D
Reactions were monitored by TLC (silica gel, 250 µm), and column
chromatographic purifications were performed on 200-300 mesh silica
gel. Solvents used for eluting the compounds, as well as TLC conditions
and Rf values, are provided under individual compound headings. THF
was distilled from LiAlH4 and redistilled over Na when needed. CH2-
Cl2 was distilled over CaCl2, and (iso-Pr)2NEt was distilled over CaH2.
All other reagents were obtained from commercial sources and used
1
2
3
4
5
-
DME
DME Cs2CO3
DME K3PO4
-
-
rt
-
+5.24 (c ) 0.00248, MeOH)
+5.64 (c ) 0.00248, MeOH)
+1.61 (c ) 0.00248, MeOH)
+3.63 (c ) 0.00248, MeOH)
85 °C 1 h
85 °C 1 h
85 °C 1 h
DME NaHCO3 85 °C 24 h +5.24 (c ) 0.00248, MeOH)
a Supplier-provided Certificate of Analysis reports that [R]D ) +5.20
(c ) 1, MeOH).
1
without further purification. H NMR spectra were recorded at 500
MHz, and 13C NMR spectra were recorded at 126 MHz. All were
referenced to the residual protonated solvent. 31P{1H} NMR spectra
were obtained at 202 MHz and were referenced to 85% H3PO4 as an
external standard. Chemical shifts are reported in parts per million (δ),
and coupling constants are in hertz. The sugar protons are numbered
1′-5′ beginning at the anomeric carbon and proceeding via the carbon
chain to the primary carbinol carbon.
sion of 17 to the phosphoramidite 18 was accomplished by
conventional methods (47% yield).
To demonstrate the utility of 18 for DNA modification, we
have undertaken preliminary experiments to incorporate 18 into
the 11-mer oligonucleotide 19 using standard phosphoramidite
chemistry. The support-bound reactive DNA 19 was then
exposed to a DMF solution of morpholine for 24 h at room
temperature. The DNA was subjected to standard cleavage from
the support and deprotection followed by LC/MS analysis. This
analysis showed the presence of the desired morpholine modified
DNA oligomer 20 (mass 3665.7) and the T10 oligomer (mass
2980.3) in a nearly 1:1 ratio indicating the coupling efficiency
of 18 to be ∼50%. However, it is important to note that DNA
assembly was performed with only a very small amount of 18
(6 mg). Coupling efficiency can be substantially improved by
altering factors such as increased concentrations of 18, longer
coupling times, etc. Most important is the fact that only the
desired morpholine modified oligomer 20 was obtained in the
reaction, and none of the ammonia displacement product was
observed. The latter would have been formed in the ammonia
cleavage step had incomplete displacement by morpholine
occurred. This showed that the DNA containing the reactive
nucleoside underwent conversion in essentially quantitative yield
to 20. The results from these experiments clearly indicate the
9-[2-Deoxy-3,5-bis-O-(tert-butyldimethylsilyl)-â-D-erythro-pento-
furanosyl]purin-6-yloxy-tris(dimethylamino)phosphonium Hexaflu-
orophosphate (3a). In a clean, dry reaction vial equipped with a stirring
bar were placed 3′,5′-bis-O-(tert-butyldimethylsilyl)-2′-deoxyinosine
(1a) (0.100 g, 0.208 mmol) and BOP (0.184 g, 0.416 mmol). Dry THF
(2.0 mL) and (iso-Pr)2NEt (54.3 µL, 0.312 mmol) were added, and the
mixture was allowed to stir at room temperature for 4 h and then
evaporated. The crude material was taken up in CH2Cl2 and washed
with water. The organic layer was separated, dried over Na2SO4, and
concentrated. Purification by preparative thin layer chromatography
(SiO2, 5% MeOH in CH2Cl2) afforded ∼10 mg (6% yield) of the
1
product as a clear gum. Rf (5% MeOH in CH2Cl2) ) 0.26. H NMR
(500 MHz, CDCl3): δ 8.68 (s, 1H, Ar-H), 8.50 (s, 1H, Ar-H), 6.54
(t, 1H, H-1′, J ) 6.4), 4.63 (m, 1H, H-3′), 4.06 (q, 1H, H-4′, J )
3.3), 3.89 (dd, 1H, H-5′, J ) 11.2, 3.7), 3.79 (dd, 1H, H-5′, J )
11.2, 2.9), 2.66 (app quint, 1H, H-2′, Japp ∼ 6.4), 2.52 (ddd, 1H, H-2′,
J ) 13.2, 5.9, 3.9), 2.92 (d, 18H, NCH3, JH-P ) 11.2), 0.92, 0.90 (2s,
(18) Ferentz, A. E.; Verdine, G. L. Nucleosides Nucleotides 1992, 11, 1749-
1763.
(19) Bae, S.; Lakshman, M. K. Provisional Patent Application 60/829,168, 2006.
9
J. AM. CHEM. SOC. VOL. 129, NO. 4, 2007 787

A R T I C L E S
Bae and Lakshman
Scheme 6. Synthesis of a DNA Building Block and Site-Specific DNA Modification
18H, tert-Bu), 0.11, 0.09, 0.08 (3s, 12H, SiCH3). 13C NMR (126 MHz,
CDCl3): δ 154.4, 152.7, 151.5, 144.5, 123.0, 88.4, 85.2, 71.9, 62.7,
41.4, 37.3 (d, JC-P ) 4.5), 25.9, 25.7, 18.4, 18.0, -4.7, -4.8, -5.4,
-5.5. 31P{1H} NMR (202 MHz, CDCl3): δ 35.44 (s, P[N(CH3)2]3),
-143.31 (septet, PF6, JP-F ) 712).
120.0, 108.6, 88.9, 85.5, 76.4, 71.6, 62.2, 26.1, 25.8, 25.6, 18.5, 18.0,
17.8, -4.4, -4.7, -4.73, -5.0, -5.3, -5.4. FAB HRMS calcd for
C34H58N7O5Si3 (M+ + H) 728.3807, found 728.3818.
O6-(Benzotriazol-1-yl)-2′-deoxyinosine (5c). In a 50 mL round-
bottom flask equipped with a stirring bar were placed 2′-deoxyinosine
1c (0.504 g, 2.00 mmol) and BOP (1.770 g, 4.00 mmol). DMF (20
mL) and (iso-Pr)2NEt (0.70 mL, 4.00 mmol) were added, and the
mixture was allowed to stir at room temperature for 26 h. The reaction
mixture was evaporated with toluene several times. The crude product
was dissolved in EtOAc and washed with water. The organic layer
was separated, dried over Na2SO4, and concentrated. Chromatographic
purification (SiO2, elution with 10% MeOH in CH2Cl2) afforded 0.421
g (57% yield) of compound 5c as a pale brownish-white foam. Rf (10%
O6-(Benzotriazol-1-yl)-3′,5′-bis-O-(tert-butyldimethylsilyl)-2′-deoxy-
inosine (5a). In a 50 mL round-bottom flask equipped with a stirring
bar were placed 3′,5′-bis-O-(tert-butyldimethylsilyl)-2′-deoxyinosine
(1a) (1.005 g, 2.090 mmol) and BOP (1.850 g, 4.181 mmol). THF (20
mL) and (iso-Pr)2NEt (0.55 mL, 3.135 mmol) were added, and the
mixture was allowed to stir at room temperature for 23 h. The mixture
was evaporated, and CH2Cl2 (50 mL) was added. The mixture was
washed with water, dried over Na2SO4, and concentrated. Chromato-
graphic purification (SiO2, elution with 20% EtOAc in hexanes)
afforded 1.038 g (83% yield) of compound 5a as a white foam. Rf
(20% EtOAc in hexanes) ) 0.14. 1H NMR (500 MHz, CDCl3): δ 8.54
(s, 1H, Ar-H), 8.41 (s, 1H, Ar-H), 8.14 (d, 1H, Ar-H, J ) 8.3),
7.55-7.41 (m, 3H, Ar-H), 6.55 (t, 1H, H-1′, J ) 6.1), 4.65 (m, 1H,
H-3′), 4.06 (q, 1H, H-4′, J ) 3.1), 3.92 (dd, 1H, H-5′, J ) 11.2,
3.9), 3.80 (dd, 1H, H-5′, J ) 11.2, 2.7), 2.65 (app quint, 1H, H-2′,
Japp ∼ 6.3), 2.52 (m, 1H, H-2′), 0.93, 0.92 (2s, 18H, tert-Bu), 0.11 (s,
12H, SiCH3). 13C NMR (126 MHz, CDCl3): δ 159.0, 153.6, 151.3,
143.5, 128.9, 128.7, 124.8, 120.6, 120.0, 108.6, 88.2, 85.0, 71.6, 62.6,
41.7, 25.9, 25.7, 18.4, 18.0, -4.7, -4.8, -5.4, -5.5. FAB HRMS calcd
for C28H44N7O4Si2 (M+ + H) 598.2993, found 598.2986.
1
MeOH in CH2Cl2) ) 0.38. H NMR (500 MHz, CDCl3): δ 8.43 (s,
1H, Ar-H), 8.24 (s, 1H, Ar-H), 8.16-8.14 (m, 1H, Ar-H), 7.57-
7.54 (m, 1H, Ar-H), 7.49-7.46 (m, 1H, Ar-H), 6.46 (dd, 1H, H-1′,
J ) 9.2, 5.5), 5.10 (dd, 1H, OH, J ) 11.3, 2.3, CD3OD exchangeable),
4.83 (br m, 1H, H-3′), 4.25 (br s, 1H, H-4′), 3.97 (dt, 1H, H-5′,
J ) 12.8, 2.0), 3.82 (app td, 1H, H-5′, Japp ∼ 11.9, 1.8), 3.07 (ddd,
1H, H-2′, J ) 13.4, 9.5, 4.9), 2.42 (dd, 1H, H-2′, J ) 13.6, 5.7),
2.05 (d, 1H, OH, CD3OD exchangeable). 13C NMR (126 MHz,
CDCl3): δ 159.1, 153.1, 151.1, 144.8, 143.2, 129.1, 128.8, 125.2, 120.5,
120.3, 108.7, 88.9, 86.9, 72.2, 62.7, 40.9. FAB HRMS calcd for
C16H16N7O4 (M+ + H) 370.1264, found 370.1258.
O6-(Benzotriazol-1-yl)inosine (5d). In a reaction vial equipped with
a stirring bar were placed inosine 1d (0.100 g, 0.373 mmol) and BOP
(0.330 g, 0.746 mmol). DMF (3.7 mL) and (iso-Pr)2NEt (97.3 µL, 0.559
mmol) were added. The reaction vial was flushed with N2, and the
mixture was allowed to stir at room temperature for 68 h. Water was
added to the reaction mixture, and the mixture was extracted with CH2-
Cl2. The organic layer was separated, washed with brine, dried over
Na2SO4, and concentrated. Because the crude product contained residual
DMF, toluene was added and evaporated several times. Chromato-
graphic purification (SiO2, elution with 5% MeOH in CH2Cl2) afforded
76 mg (53% yield) of 5d as a white powder. Rf (5% MeOH in CH2Cl2)
) 0.09. 1H NMR (500 MHz, DMSO-d6): δ 8.95 (s, 1H, Ar-H), 8.53
(s, 1H, Ar-H), 8.21 (d, 1H, Ar-H, J ) 8.8), 7.80 (d, 1H, Ar-H, J )
8.3), 7.66 (t, 1H, Ar-H, J ) 7.6), 7.56 (t, 1H, Ar-H, J ) 7.6), 6.09
(d, 1H, H-1′, J ) 5.4), 5.55 (d, 1H, OH, J ) 5.9, D2O exchangeable),
5.24 (d, 1H, OH, J ) 5.4, D2O exchangeable), 5.07 (t, 1H, OH, J )
O6-(Benzotriazol-1-yl)-2′,3′,5′-tris-O-(tert-butyldimethylsilyl)-
inosine (5b). As described for the synthesis of 5a, 5b was prepared by
a reaction of 2′,3′,5′-tris-O-(tert-butyldimethylsilyl)inosine 1b (3.055
g, 5.00 mmol), BOP (4.425 g, 10.0 mmol), and (iso-Pr)2NEt (1.31 mL,
7.5 mmol) in dry THF (50.0 mL) at room temperature over 43 h.
Chromatographic purification (SiO2, elution with 20% EtOAc in
hexanes) afforded 2.898 g (80% yield) of compound 5b as a white,
foamy solid. Rf (20% EtOAc in hexanes) ) 0.32. 1H NMR (500 MHz,
CDCl3): δ 8.63 (s, 1H, Ar-H), 8.40 (s, 1H, Ar-H), 8.15 (d, 1H, Ar-
H, J ) 8.3), 7.56-7.45 (m, 3H, Ar-H), 6.16 (d, 1H, H-1′, J ) 4.4),
4.58 (t, 1H, H-2′, J ) 4.4), 4.34 (t, 1H, H-3′, J ) 4.2), 4.18 (app q,
1H, H-4′, Japp ∼ 3.1), 4.06 (dd, 1H, H-5′, J ) 11.7, 3.4), 3.82 (dd,
1H, H-5′, J ) 11.7, 2.4), 0.98, 0.94, 0.82 (3s, 27H, tert-Bu), 0.17,
0.16, 0.11, 0.10, 0.0, -0.17 (6s, 18H, SiCH3). 13C NMR (126 MHz,
CDCl3): δ 159.0, 153.9, 151.4, 143.9, 143.5, 129.0, 128.7, 124.8, 120.6,
9
788 J. AM. CHEM. SOC. VOL. 129, NO. 4, 2007

O6-(Benzotriazol-1-yl)inosine Derivatives
A R T I C L E S
5.6, D2O exchangeable), 4.62 (q, 1H, H-2′, J ) 5.4), 4.21 (q, 1H,
H-3′, J ) 4.7), 4.01 (m, 1H, H-4′), 3.71 (m, 1H, H-5′), 3.60 (m,
1H, H-5′). 13C NMR (126 MHz, DMSO-d6): δ 159.0, 154.8, 151.9,
146.1, 143.5, 130.1, 129.3, 126.1, 120.7, 119.8, 110.2, 88.9, 86.5, 74.7,
70.9, 61.8. FAB HRMS calcd for C16H16N7O5 (M+ + H) 386.1213,
found 386.1191.
(benzotriazol-1-yl)-5′-O-(4,4′-dimethoxytrityl)-2′-deoxyinosine (17) (0.030
g, 0.0447 mmol), and the vial was transferred to a glove bag maintained
under N2. Dry CH2Cl2 (0.5 mL), (iso-Pr)2NEt (19.5 µL, 0.112 mmol),
and 2-cyanoethyl diisopropylchlorophosphoramidite (19.9 µL, 0.0894
mmol) were added. The mixture was removed from the glove bag and
allowed to stir at room temperature for 1 h under a N2 atmosphere.
The reaction mixture was diluted with CH2Cl2 and washed with
saturated NaHCO3 followed by brine. The organic layer was dried over
Na2SO4 and concentrated. Chromatographic purification (SiO2 packed
with 50:50:1 EtOAc/hexanes/(iso-Pr)2NEt and eluted with 50% EtOAc
in hexanes) afforded 18.4 mg (47% yield) of compound 18 as a white,
foamy solid. Rf (50:50:1 EtOAc/hexanes/(iso-Pr)2NEt) ) 0.51 and 0.45
for the two phosphoramidite diastereomers. 31P{1H} NMR (202 MHz,
CDCl3): δ 150.08 and 150.00.
O6-(Benzotriazol-1-yl)-5′-O-(4,4′-dimethoxytrityl)-2′-deoxy-
inosine (17). In a 100 mL round-bottom flask equipped with a stirring
bar were placed 5′-O-(4,4′-dimethoxytrityl)-2′-deoxyinosine17 (16) (0.7
g, 1.262 mmol), BOP (1.117 g, 2.524 mmol), (iso-Pr)2NEt (0.44 mL,
2.524 mmol), and dry THF (50.0 mL). The mixture was allowed to
stir at room temperature, under a N2 balloon, for 40 h. The reaction
mixture was evaporated, diluted with EtOAc (200 mL), and washed
with brine. The organic layer was dried over Na2SO4 and concentrated.
Chromatographic purification (SiO2, packed with 1% (iso-Pr)2NEt in
EtOAc and elution with EtOAc) afforded 0.718 g (85% yield) of
compound 17 as a white foam. Rf (EtOAc) ) 0.31. 1H NMR (500 MHz,
CDCl3): δ 8.31 (s, 1H, Ar-H), 8.28 (s, 1H, Ar-H), 8.13 (d, 1H, Ar-
H, J ) 8.2), 7.55-7.45 (m, 3H, Ar-H), 7.38 (d, 2H, Ar-H, J ) 7.3),
7.29-7.19 (m, 7H, Ar-H), 6.80 (d, 4H, Ar-H, J ) 8.9), 6.52 (t, 1H,
H-1′, J ) 6.4), 4.73 (br m, 1H, H-3′), 4.20 (q, 1H, H-4′, J ) 4.2),
3.77 (s, 6H, OCH3), 3.45 (dd of ABquartet, 1H, H-5′, J ) 10.4, 4.6),
3.40 (dd of ABquartet, 1H, H-5′, J ) 10.4, 5.0), 2.87 (app quint, 1H,
H-2′, Japp ∼ 6.6), 2.62 (ddd, 1H, H-2′, J ) 13.4, 6.2, 4.3), 2.39 (d,
1H, 3′-OH, J ) 3.1). 13C NMR (126 MHz, CDCl3): δ 159.1, 158.6,
153.5, 151.4, 144.4, 143.5, 135.5, 135.4, 130.0, 128.9, 128.8, 128.0,
127.9, 127.1, 124.8, 120.6, 120.1, 113.2, 108.7, 86.8, 86.3, 85.0, 72.6,
63.5, 55.2, 40.3. FAB HRMS calcd for C37H34N7O6 (M+ + H) 672.2571,
found 672.2583.
Acknowledgment. This work was partially supported by NSF
grant CHE-0314326 and a PSC CUNY-37 award. Purchase of
a mass spectrometer was funded by NSF grant CHE-0520963.
We are very grateful to Dr. Surendra Chaturvedi (Primesyn Labs
Inc., NJ) for the DNA assembly and postsynthetic DNA
modification experiment with 18 and to Dr. Bill Boggess and
Ms. Nonka Sevova (University of Notre Dame mass spectrom-
etry facility) for high-resolution mass spectral analyses.
Supporting Information Available: Synthetic procedures and
spectral data not described in the manuscript as well as 1H NMR
spectra of 3a, 5a-d, 7a-o, 8a-g, 9a,b, 10a,b, 13-15, 17, and
18 and the 31P{1H} NMR spectrum of 18. This material is
available free of charge via the Internet at http://pubs.acs.org.
O6-(Benzotriazol-1-yl)-3′-O-[(N,N-diisopropylamino)(â-cyanoeth-
oxy)phosphinyl]-5′-O-(4,4′-dimethoxytrityl)-2′-deoxyinosine (18). In
a clean, dry vial equipped with a stirring bar was placed O6-
JA064682N
9
J. AM. CHEM. SOC. VOL. 129, NO. 4, 2007 789