Synthesis and properties of DNA containing cyclonucleosides

Article abstract of DOI:10.1080/15257770.2012.716189

Here, we present efficient syntheses of the R and S diastereomers of 8,5′-cyclo-2′-deoxyadenosine and 6,5′-cyclo-2′- deoxyuridine. We incorporated these interesting nucleosides into DNA to study how the cyclo linkage affects the stability of duplex formation.

Full text of DOI:10.1080/15257770.2012.716189

This article was downloaded by: [Washington University in St Louis]
On: 16 June 2013, At: 19:21
Publisher: Taylor & Francis
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered
office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK
Nucleosides, Nucleotides and Nucleic
Acids
Publication details, including instructions for authors and
subscription information:
http://www.tandfonline.com/loi/lncn20
Synthesis and Properties of DNA
Containing Cyclonucleosides
Han Yueh a , Hongchuan Yu a , Christopher S. Theile a , Ayan Pal a
,
Allen Horhota a , Nicholas Greco a , Carl V. Christianson a & Larry W.
McLaughlin a
a Department of Chemistry, Boston College, Merkert Chemistry
Center, Chestnut Hill, Massachusetts, USA
To cite this article: Han Yueh , Hongchuan Yu , Christopher S. Theile , Ayan Pal , Allen Horhota ,
Nicholas Greco , Carl V. Christianson & Larry W. McLaughlin (2012): Synthesis and Properties of DNA
Containing Cyclonucleosides, Nucleosides, Nucleotides and Nucleic Acids, 31:9, 661-679
To link to this article: http://dx.doi.org/10.1080/15257770.2012.716189
PLEASE SCROLL DOWN FOR ARTICLE
Full terms and conditions of use: http://www.tandfonline.com/page/terms-and-conditions
This article may be used for research, teaching, and private study purposes. Any
substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing,
systematic supply, or distribution in any form to anyone is expressly forbidden.
The publisher does not give any warranty express or implied or make any representation
that the contents will be complete or accurate or up to date. The accuracy of any
instructions, formulae, and drug doses should be independently verified with primary
sources. The publisher shall not be liable for any loss, actions, claims, proceedings,
demand, or costs or damages whatsoever or howsoever caused arising directly or
indirectly in connection with or arising out of the use of this material.

Nucleosides, Nucleotides and Nucleic Acids, 31:661–679, 2012
Copyright ꢀC Taylor and Francis Group, LLC
ISSN: 1525-7770 print / 1532-2335 online
DOI: 10.1080/15257770.2012.716189
SYNTHESIS AND PROPERTIES OF DNA CONTAINING
CYCLONUCLEOSIDES
Han Yueh, Hongchuan Yu, Christopher S. Theile, Ayan Pal, Allen Horhota,
Nicholas Greco, Carl V. Christianson, and Larry W. McLaughlin
Department of Chemistry, Boston College, Merkert Chemistry Center, Chestnut Hill,
Massachusetts, USA
Here, we present efficient syntheses of the R and S diastereomers of 8,5ꢁ-cyclo-2ꢁ-deoxyadenosine
2
and 6,5ꢁ-cyclo-2ꢁ-deoxyuridine. We incorporated these interesting nucleosides into DNA to study
how the cyclo linkage affects the stability of duplex formation.
Keywords Cyclonucleosides; 8-5ꢁ-cyclo-2ꢁ-deoxyadensine; 6,5ꢁ-cyclo-2ꢁ-deoxyuridine;
TATA binding protein
INTRODUCTION
DNA damage is a ubiquitous part of the cellular life cycle. Oxidative stress
is particularly problematic for nucleic acids. In many cases, this stress is initi-
ated by a free radical, commonly the hydroxyl radical (·OH). Hydroxyl radi-
cals are generated by both exogenous agents such as carcinogenic molecules
and ionization radiation, as well as endogenous sources.[1] DNA damage by
free radicals is commonly termed as “oxidatively induced damage to DNA” in
living cells. Such damage has been shown to result in mutagenesis, carcino-
genesis, and aging.[2] In terms of specific reactivity between these reactive
oxygen species and DNA, two main adducts are likely: the addition of oxygen
to the unsaturated bonds of the heterocyclic bases and abstraction of hydro-
gen atoms from the sugar backbone. H-abstraction has also been shown to
occur at the methyl group of thymine, generating thymine dimers. As a result
of these various damage pathways, a number of nucleobase and deoxyribose
Received 12 June 2012; accepted 23 July 2012.
Han Yueh and Hongchuan Xu contributed equally to the research presented in this manuscript.
The authors would like to acknowledge Bo Li and Marek Domin for their guidance in the analytical
techniques used in this manuscript.
Address correspondence to Larry W. McLaughlin, Department of Chemistry, Boston Col-
lege, Merkert Chemistry Center, 2609 Beacon Street, Chestnut Hill, MA 02467, USA. E-mail:
larry.mclaughlin@bc.edu
661

662
H. Yueh et al.
FIGURE 1 The R and S diastereomers of 8,5ꢁ-cyclo-2ꢁ-deoxyadenosine and 6,5ꢁ-cyclo-2ꢁ-deoxyuridine.
perturbations lead to single and double strand breaks, formation of abasic
sites, DNA—protein, and carbohydrate-nucleobase cross-links.[3–7]
In this study, we focus on 8,5ꢁ-cyclo-2ꢁ-deoxyadenosine and 6,5ꢁ-cyclo-2ꢁ-
deoxyuridine nucleosides (Figure 1). The 8,5ꢁ-cyclo-2ꢁ-deoxyadenosine rep-
resents a unique class of naturally occurring, helix-distorting DNA lesions
that result in a second covalent linkage between the sugar and base moi-
eties of a single nucleoside.[8] Unnatural 6,5ꢁ-cyclo-2ꢁ-deoxyuridine deriva-
tives were also investigated, as they have similar structural features compared
to the cyclo-deoxyadenosine moieties.
These lesions are initiated by H-abstraction at the C5ꢁ by the hydroxyl
radical species. This is followed by attack of the C5ꢁ-centered radical at the
C8 of the purine base, leading to C8–C5ꢁ-cyclization. Subsequent oxidation
of the resulting N7-centered radical results in the intramolecular cyclized
product with a covalent bond between the C5ꢁ and C8 positions of the
purine nucleoside.[9]
Both R and S diastereomers (Figure 1) are possible and although they
result in similar puckering and distortion, the unique position of the 5ꢁ
OH causes them to have different properties.[8,10–13] It has been shown that
the R-diastereomer is more prevalent under γ -radiation of an anaerobic
water solution of ss-DNA, whereas the R/S ratio is reversed in favor of the
S-diastereomer in ds-DNA.[14] Moreover, it is postulated that the R- and S-
diastereomers of cyclo-dA may be recognized differently during nucleoside
excision repair due to the stereospecificity of the repair enzymes. Studies
have shown that the R-diastereomer is more efficiently repaired than the S-
diastereomer.[15,16] These differences demand that each diastereomer needs
to be prepared separately, which lead to our synthetic efforts in this paper.
Cyclonucleoside lesions have major biological implications. These in-
clude blocking transcription and mutagenesis.[8] Studies have proven that
the lesions can alter the binding of transcription factors to their cognate DNA
recognition elements.[16] Cyclonucleosides can block the TATA binding pro-
tein (TBP) from associating with its target region, causing a transcriptional
roadblock, which in turn shuts off gene expression. The exact mechanism
of inhibition is not yet known.[8] TBP binds in the minor groove, causing
a major local conformational distortion.[17] One of the reasons that nature

Synthesis and Properties of DNA Containing Cyclonucleosides
663
may have chosen the TATA box instead of a GCGC box for transcriptional
control is the ease of A·T minor groove deformation rather than the stronger
G·C base-pairs.[17] Because cyclonucleosides block TBP binding, several fac-
tors may be operating; the rigidity imparted by the extra covalent bond could
deform the secondary structure of the oligonucleotide duplex, the cyclonu-
cleoside could be forming a stronger base-pair with its Watson and Crick
partner, or some combination of the two.
With stereochemically pure samples of these cyclic nucleosides, we hope
to address the question of whether or not this lesion results in stronger
hydrogen bond interactions with its Watson–Crick pair. The restricted flex-
ibility of the C5ꢁ–C8 covalent bond might help to pre-organize and thereby
reduce entropic loss while forming a base pair. On the other hand, if these
lesions destabilize the duplex formation, it will confirm that the deforma-
tion caused by the extra bond overpowers the effect of the weak A·T base
pair of the TATA box, and results in TBP inhibition. Because 6,5ꢁ-cyclo-
2ꢁ-deoxyuridine features a similar lesion, it was also synthesized to aid in
the study of duplex stability. Due to synthetic constraints, the deoxyuridine
monomer was produced instead of thymidine in order to retain aromaticity
after introducing the C6–C5ꢁ bond.
RESULTS AND DISCUSSION
Chemistry
The focus of this study was to prepare four nucleosides (two sets of di-
astereomers) based upon dA and dU. Introducing a second bond between
the sugar and the heterocycle would rigidify the nucleosides. These cyclonu-
cleosides have a second bond between C5ꢁ and C8 for the 2ꢁdeoxyadenosine
and C5ꢁ and for 2ꢁ-deoxyuridine. Model building studies conducted in our
lab suggested that cross-links at these locations would result in glycosidic
bond angles similar to those observed in double-stranded nucleic acids.
The cyclo-2ꢁ-deoxyadenosine monomer was initiated from N-benoyl-2ꢁ-
deoxyadenosine (3) with two protection steps to generate compound 5
(Scheme 1). Synthesis of 8 was carried out in three steps. Substitution of
the 5ꢁ-tosyl by iodide permitted a zinc-mediated cyclization to generate
8,5ꢁ-cyclo-2ꢁ,5ꢁ-dideoxyadenosine. Oxidation with SeO2 yielded the ketone
8, which upon reduction with NaBH4 formed the S diastereomer exclu-
sively. To obtain the R isomer, 9S was mesylated and then inverted with
hydroxide. From this point forward the two diastereomers were processed
in parallel. The newly formed 5ꢁ hydroxyls of both the R and S diastere-
omers were unable to be protected with the standard 4,4ꢁ-dimethoxytrityl
chloride (DMTrCl) protecting group due to steric interactions. Instead we
opted to use ethyl vinyl ether, which has similar deprotecting conditions as
DMTrCl, to form 1-ethylethoxy protected 11S and 11R. The ethyl vinyl ether

664
H. Yueh et al.
NHBz
NHBz
NHBz
i. NaI, Acetone (6)
N
N
N
N
N
N
ii. Zn, Pyr, then tetrachloro-
p-benzoquinone, Benzene (7)
iii. SeO2, 1,4-Dioxane
N
N
N
N
i. TsCl, Pyr (4)
ii. TIPSCl, Imidazole
AgNO3, Pyr
HO
TsO
O
N
N
O
O
O
OH
OTIPS
OTIPS
3
5
8
NaBH4,
MeOH
NHBz
NHBz
NHBz
S
i. TBAF, THF (12
ii. 2-Cyanoethyl
)
N
N
N
N
N
N
N
N
N
N,N,N',N'-tetraisopropyl
phosphoramidite,
Tetrazole, MeCN
HO
O
O
O
O
EVE, PPTS
DCM
N
N
N
O
O
O
OTIPS
O
O
OTIPS
P
N
CN
9S
11S
i. MsCl, Pyr (10)
ii. NaOH, H2O
13S
NHBz
NHBz
NHBz
R
i. TBAF, THF (12
)
ii. 2-Cyanoethyl
N,N,N',N'-tetraisopropyl
phosphoramidite,
N
N
N
N
N
N
N
N
N
N
N
N
EVE, PPTS
HO
DCM
O
O
O
O
Tetrazole, MeCN
O
O
O
O
O
OTIPS
OTIPS
P
CN
9R
11R
N
13R
SCHEME 1 Synthesis of the R and S 8,5ꢁ-cyclo-2ꢁ-deoxyadenosine phosphoramidites.
protection resulted in two additional diastereomers at the newly formed
methyl position, which co-migrated during silica gel chromatography. The
final steps involved removal of the 3ꢁ silyl protecting group and conversion
to the 3ꢁ-phosphoramidites (13S, 13R). The presence of the two phosphorus
diastereomers and the two EVE diastereomers could be confirmed by the
presence of four phosphorus resonances in the final R and S products (e.g.,
for 13R, 31P δ 147.4, 147.6, 147.7, 147.9 ppm).
The 3ꢁ,5ꢁ-Diacetyl-2ꢁ-deoxyuridine (14) was converted to the fused deriva-
tive 20 using an established protocol (Scheme 2) with the exception of the
chlorination reaction on the C5 position, which was more efficiently accom-
plished using CAN (cerium (IV) ammonium nitrate)/LiCl in 88% yield.[18]
The 3ꢁ-hydroxyl protected 6,5ꢁ-cyclo-2ꢁ,5ꢁ-dideoxyuridine, 21, was converted
to the 5ꢁ-hydroxyl compounds, 22S and 22R, using 0.5 eq. SeO2 and 1.5
eq. tBuOOH.[19] Column chromatography separated the two derivatives in
approximately 40% yield each, for an overall 80% yield.
Crystal Structure and Thermal Behavior
The four cyclonucleosides were incorporated into 12-mer or 14-mer se-
quences containing a core d(AAAA)/d(TTTT) sequence. In the purified
oligos, the presence of the cyclonucleosides had no effect on the activity of

Synthesis and Properties of DNA Containing Cyclonucleosides
665
O
N
O
O
N
i. NaI, Acetone (18)
Cl
NH
NH
NH
O
i. CAN, LiCl, AcOH, MeCN (15)
ii. 7N NH3 in MeOH (16)
iii. TsCl, Pyr then Ac2O
ii. AIBN, HSnBu3, Benzene (19)
iii. EtONa, EtOH (20)
O
iv. TIPSCl, Imidazole, AgNO3, Pyr
AcO
TsO
O
N
O
O
O
OAc
OAc
OTIPS
14
17
21
SeO2, tBuOOH,
1,4-Dioxane
i. TBAF, THF (24S)
ii. 2-Cyanoethyl
N,N,N',N'-tetraisopropyl
phosphoramidite,
Tetrazole, MeCN
O
O
N
O
N
NH
NH
NH
O
HO
O
O
O
O
EVE, PPTS, DCM
O
N
O
O
O
O
O
O
OTIPS
OTIPS
P
N
CN
22S
23S
+
25S
i. TBAF, THF (24R)
ii. 2-Cyanoethyl
N,N,N',N'-tetraisopropyl
phosphoramidite,
Tetrazole, MeCN
O
O
O
NH
NH
NH
HO
O
O
O
O
EVE, PPTS, DCM
N
O
O
N
O
O
N
O
O
O
O
OTIPS
OTIPS
P
CN
22R
N
23R
25R
SCHEME 2 Synthesis of the R and S 6,5ꢁ-cyclo-2ꢁ-deoxyuridine phosphoramidites.
nuclease enzymes, allowing us to digest the oligos and confirm the incorpo-
ration by HPLC.
After synthesizing these compounds and obtaining crystal structural in-
formation (Figure 2), it became clear that these nucleosides would not en-
hance duplex stability. The reasoning behind this is that the C5ꢁ–C8 (dA)
and C5ꢁ–C6 (dU) bonds “pull” the heterocycle away from the center of the
DNA double helix, preventing it from forming a Watson–Crick base pair.
FIGURE 2 The crystal structures of 10R, with protecting groups hidden, and deprotected 24S overlayed
with 2ꢁ-dA and 2ꢁ-dU[20,21] (Color figure available online).

666
H. Yueh et al.
χ
O4'
γS
C5'
C4'
γR
O(S)
(R)O
H
FIGURE 3 An overhead view of a cyclonucleoside highlighting the relevant torsion angles (Color figure
available online).
Torsion angles can help to explain the nature of the duplexes containing
rigid nucleosides. By introducing the bond between C8 and C5ꢁ in cyclo-
2ꢁ-deoxyadenosine and C6 and C5ꢁ in cyclo-2ꢁ-deoxyuridine the glycosidic
torsion angles, χ, became fixed at −146.7◦ and −152.1◦ respectively (Figure
3). The dA-dT core in the Dickerson Dodecamer has χ values ranging from
near −80◦ to near −115◦ while the corresponding RNA sequence exhibits χ
values around −160◦. These values suggest that the fused nucleosides might
be better accommodated in an RNA A-like duplex; however, even these
RNA/DNA heteroduplexes were destabilized (see Table 1).
The torsion angle for the C4ꢁ–C5ꢁ bond (γ ) is also problematic. The γ
torsion angle prefers a value close to 70◦ for RNA and a value near 40◦ for
DNA. In these conformations, the 5ꢁO is positioned above the sugar and
angled toward the heterocycle. In the case of the modified nucleosides, that
position is occupied up by the C8–C5ꢁ bond in cyclo-2ꢁ-deoxyadenosine or
the C6–C5ꢁ bond in cyclo-2ꢁ-deoxyuridine. The fused 5ꢁOH is left in a position
normally inhabited by the C5ꢁ hydrogens. The S diastereomer results in a
γ torsion angle near −50◦ while the R produces a torsion angle of about
−170◦.
CONCLUSIONS
Our effort toward preparing pure diastereomers of nucleosides of cyclo-
2ꢁ-deoxyadenosine and cyclo-2ꢁ-deoxyuridine in appreciable yields has been
successful. We incorporated these nucleosides into DNA sequences and
found that the Tm in most of the cases is reduced. These results are in
accordance with theoretical studies performed in the labs of John Miller.[12]
Our data proves that cyclonucleoside lesions destabilize duplex formation.

Synthesis and Properties of DNA Containing Cyclonucleosides
667
TABLE 1 Thermal denaturation studies of DNA containing cyclonucleosides (cyclonucleosides
denoted by underline and bold text)
Sequence
TM
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
5ꢁ-d(C C G G A A A A C G C C)/5ꢁ-d(G G C G T T T T C C G G)
49
40
36
28
36
51
17
28
28
16
5ꢁ-d(C C G G AS A AS A C G C C)/5ꢁ-d(G G C G T T T T C C G G)
5ꢁ-d(C C G G AR A AR A C G C C)/5ꢁ-d(G G C G T T T T C C G G)
5ꢁ-d(C C G G A A A A C G C C)/5ꢁ-d(G G C G US T US T C C G G)
5ꢁ-d(C C G G A A A A C G C C)/5ꢁ-d(G G C G UR T UR T C C G G)
5ꢁ-d(C C G G A A A A A C G C C)/5ꢁ-d(G G C G T T T T T T C C G G)
5ꢁ-d(C C G G A AS AS AS AS A C G C C)/5ꢁ-d(G G C G T T T T T T C C G G)
5ꢁ-d(C C G G A AR AR AR AR A C G C C)/5ꢁ-d(G G C G T T T T T T C C G G)
5ꢁ-d(C C G G A A A A A A C G C C)/5ꢁ-d(G G C G T US US US US T C C G G)
5ꢁ-d(C C G G A A A A A A C G C C)/5ꢁ-d(G G C G T UR UR UR UR T C C G G)
5ꢁ-d(C C G G A AS AS AS AS A C G C C)/5ꢁ-d(G G C G T US US US US T C C G G)
5ꢁ-d(C C G G A AS AS AS AS A C G C C)/5ꢁ-d(G G C G T UR UR UR UR T C C G G)
5ꢁ-d(C C G G A AR AR AR AR A C G C C)/5ꢁ-d(G G C G T US US US US T C C G G)
5ꢁ-d(C C G G A AR AR AR AR A C G C C)/5ꢁ-d(G G C G T UR UR UR UR T C C G G)
5ꢁ-d(C C G G A A A A C G C C)/5ꢁ-G G C G U U U U C C G G
14
∗
∗
∗
43
29
30
5ꢁ-d(C C G G AS A AS A C G C C)/5ꢁ-G G C G U U U U C C G G
5ꢁ-d(C C G G AR A AR A C G C C)/5ꢁ-G G C G U U U U C C G G
∗= Duplex formation was not observable.
This supports the hypothesis that duplex deformation caused by this lesion
overpowers the effects of the weaker TATA box base pairing, thus inhibiting
TBP binding. Our crystal structures show that the nucleobase is “pulled”
back from its Watson and Crick position, explaining why these molecules
form weaker hydrogen bonds as a base pair. These compounds are of con-
tinued interest due to the fact that they are resistant toward enzymatic repair
and may be valid drug targets.[22]
EXPERIMENTAL
General Information
Reagents were purchased from Sigma-Aldrich, Acros, Oakwood, Glen
Research, Lancaster, MP Biomedical, Chem-Impex International, Fisher,
Chem Genes and Molekula. Flash column were performed using dynamic
˚
adsorbents silica gel (60 A, particle size 32–63 µm) and TLC monitoring with
TLC Silica Gel with F-254 Indicator (Dynamic Adsorbents). TLCs were visual-
ized by 260 nm UV light and stained by 10% sulfuric acid. All reactions were
carried out under a nitrogen atmosphere with dry solvent under anhydrous
conditions unless indicated otherwise. Dry tetrahydrofuran (THF), diethyl
ether (Et2O), N ,N-dimethylformamide (DMF), pyridine (pyr), acetonitrile
(MeCN), and dichloromethane (DCM) were obtained by passing commer-
cially available pre-dried, oxygen-free formulations through activated alu-
mina columns. Dry ethanol was purchased from Sigma–Aldrich and used

668
H. Yueh et al.
directly. NMR spectra were taken by Varian VNMRS400, VNMRS500, or IN-
OVA 500 instruments and calibrated using residual undeuterated solvent
(CDCl3: δ H = 7.24 ppm, δ C = 77.23 ppm, DMSO-d6: δ H = 2.50 ppm,
δ C = 39.51 ppm). Abbreviations of multiplicities were designated as fol-
low: s = singlet, d = doublet, t = triplet, q = quartet, m = multiplet, and
b = broad. High-resolution mass spectra (HRMS) were recorded on a waters
LCT or JEOL AccuTOF mass spectrometer using ESI (electrospray ioniza-
tion) or DART (direct analysis in real time). UV spectra were acquired by
Beckman (Fullerton, CA, USA) DU650 spectrophotometer. HPLC was per-
formed on Waters (Milford, MA, USA) Delta 600 controllers with a 2487 dual
wavelength detector. DNA oligomers were prepared using an Applied Biosys-
tems 394 DNA/RNA synthesizer with standard solid phase phosphoramidite
techniques and reagents purchased from Glenn Research (Sterling, VA,
USA). RNA oligomers were purchased from Integrated DNA Technologies
(Coralville, IA, USA). Polyacrylamide Gel Electrophoresis was performed by
using a Hoefer Scientific (San Francisco, CA, USA) apparatus. Gel imaging
was performed on Bio-Rad (Hercules, CA, USA) FXpro Molecular Imager
with ethidium bromide staining or Kodak-K radioisotope screens.
Synthesis
N-Benzoyl-5ꢀ-tosyl-2ꢀ-deoxyadenosine (4)
N-benoyl-2ꢁ-deoxyadenosine (10.2 g, 28.7 mmol) was dissolved in dry
pyridine (140 mL) and cooled to 0◦C. p-tosylsulfonyl chloride (8.20 g,
43.0 mmol) was added and the reaction mixture was allowed to warm to
ambient temperature and stir overnight. The reaction was quenched with
water at 0◦C and solvents were removed under reduced pressure. The residue
was dissolved in DCM and washed with water three times and once with brine.
The organic layer was dried over sodium sulfate and concentrated in a ro-
tary evaporator. The crude product was purified by flash chromatography
(DCM/MeOH 95:5) to yield the 4 as a white solid (12.1 g, 83%). TLC Rf =
0.32 (DCM/MeOH 90:10). 1H NMR (400 MHz; CDCl3): δ 2.43 (s, 3H), 2.49
(d, 1H), 2.58 (m, 1H), 2.97 (m, 1H), 4.23–4.3 (m, 3H), 4.83 (bm, 1H), 6.49
(t, 1H), 7.31 (d, 2H), 7.54 (t, 2H), 7.63 (t, 1H), 7.74 (d, 2H), 8.03 (dd,
2H), 8.18 (s, 1H), 8.75 (s, 1H), 8.98 (s, 1H). 13C NMR (100 MHz; CD3OD):
δ 41.6, 63.4, 72.8, 86.8, 89.9, 125.6, 129.6, 129.6, 129.9, 129.9, 134.0, 134.2,
134.3, 135.1, 144.7, 146.5, 151.3, 153.0, 153.1, 168.3. ESI-MS (pos.): 510.1455
[M+H]+ (HSMS calc. 510.1447).
N-Benzoyl-3ꢀ-triisopropylsilyl-5ꢀ-tosyl-2ꢀ-deoxyadenosine (5)
4 (5.60 g, 11.0 mmol), imidazole (2.30 g, 33.4 mmol), and silver nitrate
(3.78 g, 22.2 mmol) were dissolved in dry pyridine (56 mL) and cooled
to 0◦C. Triisopropylsilyl chloride (4.70 ml, 22.0 mmol) was added and the

Synthesis and Properties of DNA Containing Cyclonucleosides
669
reaction was warmed to ambient temperature. After stirring for two days
the reaction was quenched with water while on an ice bath. The solvents
were removed under reduced pressure and the residue dissolved in DCM
and washed with water three times and once with brine. The organic layer
was dried over sodium sulfate and concentrated under vacuum. The crude
product was purified by flash chromatography (DCM/MeOH 97:3) to yield
5 as a white foam (5.2 g, 72%). TLC Rf = 0.65 (DCM/MeOH 97:3). 1H NMR
(400 MHz; CDCl3): δ 1.07–1.12 (m, 21H), 2.42 (s, 3H), 2.47–2.51 (m, 1H),
2.93 (m, 1H), 4.22 (s, 1H), 4.19–4.3 (m, 2H), 4.73 (t, 1H), 6.47 (t, 1H), 7.29
(d, 2H), 7.52 (t, 2H), 7.6 (q, 1H), 7.74 (d, 2H), 8.03 (d, 2H), 8.20 (s, 1H),
8.73 (s, 1H). 13C NMR (100 MHz; CDCl3): δ 11.9, 17.9, 21.6, 40.3, 68.8, 73.0,
85.0, 85.4, 123.4, 125.8, 127.9, 128.8, 129.9, 132.2, 132.8, 133.4, 141.9, 145.2,
149.4, 151.4, 152.1, 164.7. ESI-MS (pos.): 666.2762 [M+H]+ (HSMS calc.
666.2782).
N-Benzoyl-3ꢀ-triisopropylsilyl-5ꢀ-iodo-2ꢀ-deoxyadenosine (6)
5 (4.26 g, 6.40 mmol) and sodium iodide (3.84 g, 2.56 mmol) were
dissolved in dry acetone (32 mL) and refluxed for four hours. The solid
suspension filtered through Celite and the filtrate was concentrated under
vacuum. The crude was purified by flash chromatography (DCM/MeOH
97.5/2.5) to yield 6 as a white foam (3.9 g, 98%) TLC Rf = 0.42 (DCM/MeOH
97.5/2.5). 1H NMR (400 MHz; CDCl3): δ 1.04–1.14 (m, 21H), 2.23 (s, 1H),
2.47 (m, 1H), 3.08 (m, 1H), 3.36–3.53 (m, 2H), 4.07 (m, 1H), 4.70 (t, 1H),
6.44 (t, 1H), 7,47 (t, 2H), 7.56 (t, 1H), 7.99 (d, 2H), 8.24 (s, 1H), 8.74 (s,
1H), 9.22 (s, 1H). 13C NMR (100 MHz; CDCl3): δ 6.3, 12.0, 18.0, 39.9, 75.4,
85.2, 86.7, 123.7, 127.8, 128.8, 132.7, 133.6, 142.2, 149.6, 151.4, 152.5, 164.6.
ESI-MS (pos.): 622.1715 [M+H]+ (HSMS calc. 622.1710).
N-Benzoyl-3ꢀ-triisopropylsilyl-8,5ꢀ-cyclo-2ꢀ,5ꢀ-dideoxyadenosine (7)
6 (1.25 g, 2.00 mmol) and zinc powder (2.63 g, 40.2 mmol) were dis-
solved in dry pyridine (70 mL) and heated to 80◦C for 4 hours. The zinc
powder was removed by filtration and the filtrate was concentrated under
reduced pressure. The residue and tetrachloro-1,4-benzoquinone (0.490 g,
2.00 mmol) were dissolved in benzene (40 mL) and refluxed for 2 hours.
The solvent was then evaporated under vacuum and the crude product was
purified by flash chromatography (DCM/MeOH 98:2) to yield the cyclized
product 7 as a yellow foam (0.42 g, 42%). TLC Rf = 0.32 (DCM/MeOH
1
97:3). H NMR (400 MHz; CDCl3): δ 0.98–1.07 (m, 21H), 2.36 (m, 1H),
2.66 (m, 1H), 3.11 (d, 1H), 3.58 (m, 1H), 4.51 (m, 1H), 4.80 (d, 1H), 6.58
(d, 1H), 7.50 (t, 2H), 7.59 (t, 1H), 7.99 (t, 2H), 8.68 (s, 1H), 8.74 (s, 1H).
13C NMR (100 MHz; CDCl3): δ 11.8, 17.8, 31.2, 48.1, 75.3, 83.2, 84.8, 127.8,
128.8, 132.8, 133.5, 146.8, 147.4, 148.1, 148.3, 151.8, 164.5. ESI-MS (pos.):
516.2425 [M+Na]+ (HSMS calc. 516.2407).

670
H. Yueh et al.
N-Benzoyl-3ꢀ-triisopropylsilyl-5ꢀ-keto-8,5ꢀ-cyclo-2ꢀ-deoxyadenosine (8)
7 (0.186 g, 0.377 mmol) and selenium dioxide (90.0 mg, 0.814 mmol)
were dissolved in 1,4-dioxane (50 mL) and refluxed for 1 hour. The reaction
mixture was filtered through Celite and the solvent removed under vacuum.
The product was purified by flash chromatography (DCM/MeOH 97:3) to
yield 8 as a yellow foam (0.170 g, 89%). TLC Rf = 0.23 (DCM/MeOH 95:5).
1H NMR (400 MHz; CDCl3): δ 1.02–1.07 (m, 21H), 2.14 (s, 1H), 2.56 (m,
1H), 2.72 (m, 1H), 4.76 (dd, 1H), 4.96 (s, 1H), 6.83 (d, 1H), 7.52 (t, 2H),
7.61 (t, 1H), 7.97 (d, 2H), 8.92 (s, 1H), 9.19 (s, 1H). 13C NMR (100 MHz;
CDCl3): δ 11.8, 17.8, 44.8, 72.7, 85.9, 92.3, 122.9, 127.9, 129.0, 133.2, 140.8,
147.9, 148.3, 152.3, 156.1, 164.3, 184.6. ESI-MS (pos.): 530.2186 [M+Na]+
(HSMS calc. 530.2200).
N-Benzoyl-3ꢀ-triisopropylsilyl-8,5ꢀ(S)-cyclo-2ꢀ-deoxyadenosine (9S)
Compound 8 (0.454 g, 0.890 mmol) was dissolved in methanol (27 mL)
and cooled to 0◦C, 0.1 M sodium borohydride aqueous solution (5 mL) was
added and the stirring reaction was allowed to warm to ambient temperature.
After 1.5 hours the reaction was neutralized with 1N hydrochloric acid and
the solvents were removed under vacuum. The crude product was purified
by flash chromatography (DCM/MeOH 95:5) to yield 9S as a white powder
1
(0.425 g, 93%). TLC Rf = 0.22 (DCM/MeOH 97:3). H NMR (400 MHz;
CDCl3): δ 0.95–1.03 (m, 21H), 2.31 (m, 1H), 2.55 (s, 1H), 4.56 (d, 1H),
4.88 (q, 1H), 5.03 (s, 1H), 5.46 (d, 1H), 6.50 (d, 1H), 7.50 (t, 2H), 7.59 (t,
1H), 7.97 (d, 2H), 8.69 (s, 1H), 8.93 (s, 1H). 13C NMR (100 MHz; CDCl3): δ
11.8, 17.9, 47.0, 64.8, 69.3, 85.5, 86.4, 122.6, 127.8, 128.9, 132.9, 133.4, 148.6,
149.9, 151.5, 152.5, 164.8. ESI-MS (pos.): 510.2534 [M+H]+ (HSMS calc.
510.2537).
N-Benzoyl-3ꢀ-triisopropylsilyl-5ꢀ-mesyl-8,5ꢀ(S)-cyclo-2ꢀ-deoxyadenosine (10)
Compound 9S (0.570 g, 0.570 mmol) was dissolved in pyridine (5.7 mL)
and methanesulfonyl chloride (110 µL, 1.43 mmol) was added at 0◦C. The
reaction was slowly warmed to ambient temperature and stirred overnight be-
fore being quenched with water. The solvents were removed under vacuum
and the residue was dissolved in DCM and washed with water three times
and once with brine. After removing the DCM under reduced pressure, the
crude product was purified by flash chromatography (DCM/MeOH 97.5:2.5)
to yield the mesylated product 10 as a white powder (0.31 g, 91%). TLC Rf
1
= 0.38 (DCM/MeOH 97.5:2.5). H NMR (400 MHz; CDCl3): δ 1.01–1.10
(m, 21H), 2.40 (m, 1H), 2.66 (m, 1H), 3.57 (s, 3H), 4.89–4.93 (m, 2H), 6.12
(d, 1H), 6.55 (d, 1H), 7.52 (t, 2H), 7.61 (t, 1H), 7.94 (d, 2H), 8.68 (s, 1H),
8.73 (s, 1H). 13C NMR (100 MHz; CDCl3): δ 11.8, 17.9, 40.0, 46.7, 70.1, 70.4,
85.1, 85.5, 122.8, 127.8, 129.0, 133.0, 133.4, 145.2, 149.2, 150.0, 152.9, 164.5.
ESI-MS (pos.): 610.2136 [M+Na]+ (HSMS calc. 610.2132).

Synthesis and Properties of DNA Containing Cyclonucleosides
671
N-Benzoyl-3ꢀ-triisopropylsilyl-8,5ꢀ(R)-cyclo-2ꢀ-deoxyadenosine (9R)
Compound 10 (0.370 g, 0.630 mmol) was dissolved in a DMF/water
mixture (50:17.6 mL) and 0.1 M sodium hydroxide (7.55 mL) was added
over 5 hours and after complete addition stirred for an additional hour.
The reaction mixture was neutralized with 1 N hydrochloric acid and the
solvents were removed under vacuum. The crude product was purified by
flash chromatography (DCM/MeOH 97:3) to yield 9R as a white powder
(0.26 g, 80%). TLC Rf = 0.17 (DCM/methanol = 30/1). 1H NMR (400 MHz;
CDCl3): δ 1.01–1.08 (m, 21H), 2.35 (m, 1H), 2.54 (m, 1H), 4.42 (q, 1H),
4.67 (s, 1H), 4.94 (d, 1H), 6.19 (s, 1H), 6.61 (d, 1H), 7.53 (dt, 2H), 7.59 (dt,
1H), 8.07 (dd, 2H), 8.75 (s, 1H), 9.32 (s, 1H). 13C NMR (100 MHz; CDCl3): δ
11.8, 17.9, 45.9, 65.8, 71.7, 85.2, 89.0, 121.6, 128.2, 128.8, 132.9, 133.5, 148.0,
149.1, 149.5, 153.1, 165.0. ESI-MS (pos.): 532.2351 [M+Na]+ (HSMS calc.
532.2356).
N-Benzoyl-3ꢀ-triisopropylsilyl-5ꢀ-(1-ethoxyethyl)-8,5ꢀ(S)-cyclo-2ꢀ-
deoxyadenosine (11S) and N-Benzoyl-3ꢀ-triisopropylsilyl-5ꢀ-(1-ethoxyethyl)-
8,5ꢀ(R)-cyclo-2ꢀ-deoxyadenosine (11R)
Compound 9S (0.150 g, 0.300 mmol) and pyridinium-p-toluenesulfonate
(75.0 mg, 0.300 mmol) were dissolved in DCM (3 mL) followed by the addi-
tion of ethyl vinyl ether (0.175 mL, 1.80 mmol) and stirred overnight at am-
bient temperature. The solvents were removed under vacuum and the crude
product was purified by flash chromatography (DCM/MeOH/Et3N 96:3:1)
to yield 11S as a yellow foam (0.12 g, 86%). TLC Rf = 0.3 (DCM/MeOH
1
97:3). H NMR (400 MHz; CDCl3): δ 1.04–1.26 (m, 21H), 1.25–1.30 (bm,
3H), 1.50 (dd, 3H, J = 40.2, 5.4 Hz), 2.34 (m, 1H), 2.60 (m, 1H), 3.62–3.88
(bm, 2H), 4.74 (dd, 1H, J = 18.0, 6.4 Hz), 4.95 (m, 1H), 5.21–5.56 (dq, 1H,
J = 132.4, 5.2 Hz), 5.33 (dd, 1H, J = 14.4, 6.0 Hz), 6.52 (q, 1H), 7.52 (m,
2H), 7.61 (dt, 1H), 7.97 (m, 2H), 8.74 (s, 1H). ESI-MS (pos.): 582.3085
[M+H]+ (HSMS calc. 582.3112).
11R was prepared using the same procedure as 11S with the exception
of chromatographic conditions (DCM/MeOH/Et3N 96.4/2.5/1) to yield
the product as a yellow foam (0.12 g, 90%). TLC Rf = 0.45 (DCM/MeOH
1
97:3). H NMR (400 MHz; CDCl3): δ 1.01–1.08 (m, 21H), 1.21–1.28 (bm,
3H), 1.40 (dd, 3H, J = 60.8, 5.6 Hz), 2.32 (m, 1H), 2.48 (m, 1H), 3.59–3.77
(bm, 2H), 4.38 (q, 1H), 4.82 (d, 1H), 4.87 (s, 1H), 5.14–5.44 (dq, 1H, J =
109.6, 5.6 Hz), 6.60 (d, 1H), 7.52 (dt, 2H), 7.60 (dt, 1H), 7.97 (dd, 2H),
8.76 (s, 1H), 8.86 (d, 1H). ESI-MS (pos.): 604.2913 [M+Na]+ (HSMS calc.
604.2931).
N-Benzoyl -5ꢀ-(1-ethoxyethyl)-8,5ꢀ(S)-cyclo-2ꢀ-deoxyadenosine (12S) and
N-Benzoyl-5ꢀ-(1-ethoxyethyl)-8,5ꢀ(R)-cyclo-2ꢀ-deoxyadenosine (12R)
Tetra-n-butylammonium fluoride (1.0 M solution in THF, 0.50 mL) was
added to a solution of 11S (0.132 g, 0.227 mmol) in THF (1.5 mL) at am-
bient temperature and stirred for 1 hour. The solvent was removed under

672
H. Yueh et al.
reduced pressure and the crude product was purified by flash chromatog-
raphy (DCM/MeOH/Et3N 96.5:2.5:1) to yield the 12S as a white powder
in quantitative yield (95 mg). TLC Rf = 0.22 (DCM/MeOH 97:3). 1H NMR
(400 MHz; CDCl3): δ 1.07–1.27 (dt, 3H), 1.25–1.48 (dd, 3H), 2.32 (m, 1H),
2.59–2.66 (bm, 3H), 3.62 (m, 1H), 3.81 (m, 1H), 4.84 (d, 1H), 4.93 (q, 1H),
5.28 (d, 1H), 5.33 (q, 1H), 6.50 (d, 1H), 7.52 (dt, 2H), 7.60 (dt, 1H), 7.97 (dd,
2H), 8.74 (s, 1H). ESI-MS (pos.): 426.1780 [M+H]+ (HSMS calc. 426.1777).
12R was prepared using the same procedure as the S diastereomer to yield
the product as a white powder in quantitative yield (92 mg). TLC Rf = 0.17
(DCM/MeOH 97.5:2.5). 1H NMR (400 MHz; CDCl3): δ 1.19–1.30 (bm, 3H),
1.41 (dd, 3H, J = 47.6, 5.6 Hz), 2.28 (m, 1H), 2.49 (m, 1H), 2.96 (q, 2H),
3.59–3.78 (bm, 2H), 4.42 (m, 1H), 4.86 (s, 1H), 4.87 (s, 1H), 5.14–5.36
(dq, 1H, J = 82.8, 5.6 Hz), 6.59 (d, 1H), 7.50 (dt, 2H), 7.59 (t, 1H), 7.97
(d, 2H), 8.73 (s, 1H). ESI-MS (pos.): 426.1772 [M+H]+ (HSMS calc. 426.
1777).
N-Benzoyl-3ꢀ-(2-cyanoethyl N,N-diisopropyl phosphorodiamidite)-5ꢀ-
(1-ethoxyethyl)-8,5ꢀ(S)-cyclo-2ꢀ-deoxyadenosine (13S) and
N-Benzoyl-3ꢀ-(2-cyanoethyl N,N-diisopropyl phosphorodiamidite)
-5ꢀ-(1-ethoxyethyl)-8,5ꢀ(R)-cyclo-2ꢀ-deoxyadenosine (13R)
Compound 12S (66.0 mg, 0.155 mmol) was dissolved in acetonitrile
(6.2 mL). 1H-tetrazole in acetonitrile (0.45 M solution in MeCN, 0.21 mL)
was added followed by 2-Cyanoethyl N,N,Nꢁ,Nꢁ-tetraisopropyl phosphorodi-
amidite (61.0 µL, 0.186 mmol) and the mixture was stirred overnight at
ambient temperature. The reaction was quenched with 5% trietheylamine
in methanol and the solvents were removed under vacuum. The residue
was dissolved in DCM and washed with saturated sodium bicarbonate three
times and once with brine. The organic layer was dried over sodium sulfate.
After removing the sodium sulfate, hexanes were added to effect precipi-
tation. The precipitate was collected by filtration and washed with hexanes
several times to yield 13S as a white powder (60 mg, 60%). TLC Rf = 0.39
(DCM/MeOH 95:5). 31P NMR (162 MHz; CDCl3): δ 146.7, 146.8, 147.0,
147.2. ESI-MS (pos.): 648.2671 [M+Na]+ (HSMS calc. 648.2675). Com-
pound 13R was obtained by a similar procedure as 13S to yield a white pow-
der (22 mg, 50%). TLC Rf = 0.42 (DCM/MeOH 95:5). 31P NMR (162 MHz;
CDCl3): δ 147.4, 147.6, 147.7, 147.9. ESI-MS (pos.): 648.2665 [M+Na]+
(HSMS calc. 648.2675).
5-Chloro-3ꢀ,5ꢀ-diacetyl-2ꢀ-deoxyuridine (15)
Compound 14 (34.2 g, 109 mmol) was dissolved in a acetonitrile acetic
acid mixture (1:1, 600 mL total) and cerium (IV) ammonium nitrate (120 g,
220 mmol) and lithium chloride (5.64 g, 133 mmol) were added at ambient
temperature, then slowly heated to 80◦C and kept for 6 hours. When TLC

Synthesis and Properties of DNA Containing Cyclonucleosides
673
monitoring showed the reaction was complete the reaction was quenched
with water and the solvent was removed under reduced pressure. The orange
residue was taken up in ethyl acetate (400 mL), washed with water (3 ×
150 mL) and brine (150 mL). After drying over anhydrous Na2SO4, the
organic layer was concentrated under reduced pressure and purified by
flash chromatography (DCM/Acetone 90:10), to yield 15 (37.6 g, 87.9%) as
a white foam: Rf = 0.34 (DCM/Acetone 90:10). 1H NMR (400 MHz, CDCl3)
δ 2.10 (s, 3H), 2.15 (s, 3H), 2.19 (s, 1H), 2.51 (m, 1H), 4.28 (m, 1H), 4.39
(m, 2H), 5.22 (d, J = 4.0 Hz, 1H), 6.28 (t, J = 6.0 Hz 1H), 7.78 (s, 1H),
8.86 (s, 1H). 13C NMR (100 MHz, CDCl3), δ 21.0, 21.1, 38.1, 64.0, 74.3,
82.6, 85.5, 103.2, 139.0, 150.2, 162.8, 170.4, 170.6. HRMS (ESI) Calcd for
C13H16ClN2O7, 347.06460; Found, 347.06459.
5-Chloro -2ꢀ-deoxyuridine (16)
15 (19.6 g, 56.6 mmol) was dissolved in methanol (80 mL) and 7 N
ammonia in methanol (80 mL) was added at room temperature. After stir-
ring for four hours the solvent was removed under reduced pressure and
the resulting brown oil was purified by flash chromatography (DCM/MeOH
90:10) to yield 16 (14.1 g, 94.8%): Rf = 0.33 (DCM/MeOH 90:10). The
spectral data obtained for this compound was consistent with that reported
by Suzuki.[18]
5-Chloro-3ꢀ-acetyl-5ꢀ-tosyl-2ꢀ-deoxyuridine (17)
16 (19.6 g, 74.7 mmol) was dissolved in pyridine (500 mL) and cooled to
0◦C. A solution of tosyl chloride (2.53 M in dry pyridine, 40 mL, 101 mmol)
was added drop wise over 45 minutes. The reaction mixture was slowly
warmed to ambient temperature and then stirred overnight. To the reaction
mixture was added acetic anhydride (17.5 mL, 186 mmol), and the mixture
stirred at ambient temperature for 4 hours. The reaction was quenched with
water and the volatiles were removed under reduced pressure and the brown
oil-like residue was purified by flash chromatography (DCM/Acetone 90:10)
to yield 17 as a white powder (20.9 g, 61.2%): Rf = 0.42 (DCM/Acetone
90:10). 1H NMR (CDCl3, 400 MHz) δ 2.08 (s, 3H), 2.10–2.19 (m, 1H), 2.44
(s, 3H), 2.47–2.49 (m, 1H), 4.19 (d, J = 2.0 Hz, 1H), 4.23–4.39 (m, 2H),
5.21 (d, J = 6.8 Hz, 1H), 6.29 (q, J = 4.8 Hz, 1H), 7.37 (d, J = 8.4 Hz, 2H),
7.79 (s, 1H), 7.81 (d, J = 8.0 Hz, 2H); 13C NMR (CDCl3, 100 MHz), δ 21.0,
21.9, 37.7, 69.0, 74.8, 82.8, 85.5, 110.2, 128.2, 130.5, 132.3, 136.3, 145.9, 149.2,
158.4, 170.7,. HRMS (ESI) Calcd for C18H20ClN2O8S, 459.06289; Found, 459.
06250.
5-Chloro-3ꢀ-acetyl-5ꢀ-iodo-2ꢀ-deoxyuridine (18)
A mixture of 17 (1.30 g, 2.84 mmol) and sodium iodide (2.13 g,
14.2 mmol) was dissolved in anhydrous acetone and heated to reflux for

674
H. Yueh et al.
6 hours. After quenching with water, volatiles were removed under reduced
pressure. The crude residue was dissolved in ethyl acetate (30 mL) and
washed with 2M sodium thiosulfate (25 mL) and brine (25 cmL). The or-
ganic layer was dried over sodium sulfate and concentrated under vacuum
to yield 18 as a white foam. This crude material was of sufficient purity to be
carried on to the next step. Rf = 0.50 (DCM/Acetone 90:10).
6,5ꢀ-cyclo-2ꢀ,5ꢀ-Dideoxyuridine (20)
18 (11.3 g, 27.3 mmol) was dissolved in benzene (300 mL) and heated
to reflux. A mixture of Bu3SnH (6.9 mL, 26.0 mmol) and AIBN (225 mg,
1.37 mmol) in benzene (30 mL) was added dropwise over a period of 4 hours.
After addition the mixture was refluxed for an extra 45 minutes. After TLC
indicated full conversion of starting material, the volatiles were removed
under reduced pressure to yield crude 19. The resulting yellow solid was dis-
solved in dry ethanol (160 mL), and sodium ethoxide (2.70 M in ethanol)
was added slowly. The reaction mixture was heated to reflux and stirred
for 1 hour. After being cooled to room temperature, the remaining sodium
ethoxide was neutralized with HCl (1 M aqueous solution). TLC monitor-
ing demonstrated full conversion of starting materials. Some of the prod-
uct precipitated out of solution and was filtered. The remaining liquid was
concentrated in a rotary evaporator and purified by flash chromatography
(DCM/MeOH 90:10) to yield 20 (3.84 g, 66.7%) as a white powder: Rf =
0.23 (DCM/MeOH 95:5). The spectral data obtained for this compound was
consistent with that reported by Suzuki.[18]
3ꢀ-Triisopropylsilyl-6,5ꢀ-cyclo-2ꢀ,5ꢀ-dideoxyuridine (21)
The 3ꢁ-OH cyclized intermediate (2.02 g, 9.60 mmol), AgNO3 (2.55 g,
15.0 mmol), and 1H-imidazole (1.31 g, 19.3 mmol) were dissolved in pyr
(100 mL) and N ,N -diisopropylethylamine (2.48 mL, 15.0 mmol). The stir-
ring reaction mixture was heated to 60◦C and triisopropylsilyl chloride
(3.18 mL 15.1 mmol) was added. The temperature was maintained at 60◦C
and the reaction was stirred for 12 hours. When TLC had indicated the full
conversion of starting materials, the reaction was quenched with saturated
ammonium chloride. Insoluble solids were filtered out, and the filtrate was
concentrated under reduced pressure. The residue was taken by ethyl acetate
(80 mL) and washed with saturated NaHCO3 (60 mL) and brine (60 mL).
The organic layer was dried over sodium sulfate before being evaporated un-
der vacuum and purified by flash chromatography (DCM/Acetone 5–10%
Acetone gradient) to yield 21 as a white powder (2.62 g, 76.1%): Rf = 0.21
(silica gel, 5% acetone in DCM); 1H NMR (CDCl3, 400 M Hz) δ 8.05 (s, 1H),
6.40 (d, J = 5.6 Hz, 1H), 5.47 (s, 1H), 4.52 (d, J = 6.4 Hz, 1H), 4.41 (q, J =
3.6 Hz, 1H), 3.21(ddd, J 1 = 1.8 Hz, J 2 = 6.8 Hz, J 3 = 18 Hz), 2.54(m, 2H),
2.26(m, 1H), 1.01–1.10 (m, 3H), 1.03 (d, J = 4.8 Hz, 1H) ppm; 13C NMR

Synthesis and Properties of DNA Containing Cyclonucleosides
675
(CDCl3, 100 M Hz), δ 162.1, 149.3, 149.0, 101.5, 84.5, 82.6, 74.9, 46.7, 30.7,
18.1, 12.1 ppm; HRMS (ESI-TOF) Calc. Mass for C18H31N2O4Si+ [M+H]+
367.20531, found: 367.20467.
3ꢀ-Triisopropylsilyl-6,5ꢀ(S)-cyclo-2ꢀ-deoxyuridine (22S) and
3ꢀ-Triisopropylsilyl-6,5ꢀ(R)-cyclo-2ꢀ-deoxyuridine (22R)
21 (903 mg, 2.46 mmol) and SeO2 (136 mg, 1.23 mmol) were dis-
solved in 1,4-dioxane (30 mL) and tertbutyl peroxide (480 µL, 3.48 mmol).
The reaction mixture was heated to 80◦C and stirred for 3.5 hours. When
TLC monitoring showed that the starting material was fully consumed,
the reaction mixture was cooled to ambient temperature and the insolu-
ble solid was filtered. The filtrate was concentrated under reduced pres-
sure to yield a red foam. The residue was purified by flash chromatogra-
phy (DCM/Acetone/Et3N 92:5:3) to yield crude 22S and crude 22R. Sep-
arately, each crude diastereomer was further purified by flash chromatog-
raphy (DCM/Acetone 90:10) to yield white powder 22R (390 mg, 41.4%),
Rf = 0.46 (DCM/Acetone/Et3N 87:10:3) and white powder 22S (375 mg,
1
39.8%), Rf = 0.30 (DCM/Acetone/Et3N 87:10:3). Compound 22S H NMR
(DMSO-d6, 400 M Hz) δ 11.18 (bs, 1H), 6.49 (d, J = 6.4 Hz, 1H), 6.18 (d,
J = 5.2 Hz, 1H), 5.65 (s, 1H), 4.71 (q, J = 3.6 Hz, 1H), 4.67(t, J = 6 Hz,
1H), 4.31 (d, J = 6 Hz, 1H), 2.58 (dd, J 1 = 7.6 Hz, J 2 = 16 Hz, 1H), 2.00
(dd, J 1 = 4 Hz, J 2 = 8.4 Hz, 1H), 1.01–1.08 (m, 3H), 1.03 (d, J = 3.2 Hz,
18H) ppm; 13C NMR (DMSO-d6, 100 M Hz), δ 163.3, 154.6, 149.6, 100.5,
86.3, 84.7, 69.3, 63.2, 45.7, 18.2, 11.8 ppm; HRMS (ESI-TOF) Calc. Mass for
1
C18H31N2O5Si+ [M+H]+ 383.20022, found: 383.19989. Compound 22R H
NMR (DMSO-d6, 400 M Hz) δ 11.249(bs, 1H), 6.28 (d, J = 6.4 Hz, 1H), 6.22
(d, J = 5.2 Hz, 1H), 5.62 (s, 1H), 4.46 (dd, J 1 = 3.2 Hz, J 2 = 7.2 Hz, 1H),
4.23 (d, J = 8.0 Hz, 1H), 4.22 (s, 1H), 2.44–2.47 (m, 1H), 1.92–1.98 (m, 1H),
1.05–1.14 (m, 3H), 1.02 (d, J = 6 Hz, 1H) ppm; 13C NMR (DMSO-d6, 100 M
Hz), δ 163.0, 151.6, 149.4 102.7, 89.2, 83.4, 71.1, 64.2, 43.99, 17.7, 11.2 ppm;
HRMS (ESI-TOF) Calc. Mass for C18H31N2O5Si+ [M+H]+ 383.20022, found:
383.20056.
3ꢀ-Triisopropylsilyl-5ꢀ-(1-ethoxyethyl)-6,5ꢀ(S)-cyclo-2ꢀ-deoxyuridine (23S) and
3ꢀ-Triisopropylsilyl-5ꢀ-(1-ethoxyethyl)-6,5ꢀ(R)-cyclo-2ꢀ-deoxyuridine (23R)
22S (138 mg, 0.36 mmol) was dissolved in dry DCM/ethyl vinyl ether
(2:1, 21 mL total volume). Pyridinium para-toluenesulfonate (36.4 mg,
0.14 mmol) was added and the reaction mixture was stirred at room tem-
perature for 4 hours, resulting in a clear solution. Volatiles were removed
under reduced pressure and the residue was purified by flash chromatogra-
phy (silica gel pre-washed by three column volumes of DCM/Et3N 99:1
followed by three column volumes of 100% DCM. Column eluent was
DCM/Acetone 90:10) to yield 23S as a white foam (136 mg, 83.1%), Rf = 0.33

676
H. Yueh et al.
(DCM/Acetone 90:10). Compound 23R was obtained by the same procedure
as 23S to yield a white foam (130 mg, 79.7%), Rf = 0.33 (DCM/Acetone
1
90:10). Compound 23S H NMR (CDCl3, 400 M Hz) δ 8.08 (br, 1H), 6.29 (t,
J = 6 Hz, 1H), 6.01 (d, J = 1.6 Hz, 0.4H), 5.74 (d, J = 2.0 Hz, 0.6H), 4.47–4.86
(m, 5H), 3.52–3.62 (m, 2H), 2.45–2.51 (m, 1H), 2.20 (dt, J 1 = 4.4 Hz, J 2 =
13.2 Hz), 1.34–1.38(m, 3H), 1.19–1.23 (m, 3H), 1.03–1.07 (m, 21H) ppm;
13C NMR (CDCl3, 100 M Hz), δ 162.6, 162.5, 152.9, 152.5, 149.0, 147.9, 103.2,
102.1, 101.2, 100.1, 86.2, 85.7, 85.6, 84.8, 69.7, 69.7, 69.1, 66.7, 62.3, 62.1,
46.5, 46.5, 20.2, 20.2, 18.2, 18.1, 15.5, 15.5, 12.2, 12.2 ppm; HRMS (ESI-TOF)
Calc. Mass for C21H39N2O6Si+ [M+H]+ 455.25774, found: 455.25732 Com-
1
pound 23R H NMR (CDCl3, 400 M Hz) δ 8.29 (br, 1H), 6.36 (t, J = 4 Hz,
1H), 5.76 (s, 0.4H), 5.69 (s, 0.6H), 4.21–4.99 (m, 5H), 3.47–3.64 (m, 2H),
2.38–2.50 (m, 1H), 2.17–2.22 (m, 1H), 1.35–1.40 (m, 3H), 1.18–1.23 (m,
3H), 1.07–1.11 (m, 3H), 1.04 (d, J = 3.6 Hz, 18H) ppm; 13C NMR (CDCl3,
100 M Hz), δ 162.5, 162.5, 152.9, 149.4, 149.2, 149.1, 148.6, 104.9, 104.5,
99.4, 98.7, 88.2, 86.5, 85.1, 85.0, 71.5, 71.1, 68.1, 67.7, 60.5, 59.6, 45.4, 45.1,
20.3, 19.8, 18.1, 18.0, 15.4, 15.3, 12.2, 12.1 ppm; HRMS (ESI-TOF) Calc. Mass
for C21H39N2O6Si+ [M+H]+ 455.25774, found: 455.25664.
5ꢀ-(1-Ethoxyethyl)-6,5ꢀ(S)-cyclo-2ꢀ-deoxyuridine (24S) and 5ꢀ-(1-Ethoxyethyl)-
6,5ꢀ(R)-cyclo-2ꢀ-deoxyuridine (24R)
23S (184 mg, 0.41 mmol) was dissolved in THF (4 mL). TBAF (1 M
in THF, 2 mL) was added and the reaction mixture was stirred at ambient
temperature for 45 minutes. The reaction was quenched with water and
the solvents were removed under reduced pressure. The resulting brown
oil residue was purified by flash chromatography (silica gel pre-washed by
three column volumes of DCM/Et3N 99:1 followed by three column volumes
of 100% DCM. Column eluent was DCM/MeOH 95:5) to afford 24S as a
white solid (120 mg, 0.40 mmol), Rf = 0.24, 0.20 (DCM/MeOH 95:5). 24R
was obtained by the same procedure and yielded a white solid in nearly
quantitative yield, Rf = 0.13 (DCM/MeOH 95:5). 24S: 1H NMR (DMSO-d6,
400 MHz) δ 1.14 (dt, J = 7.2, 1.6 Hz, 3H), 1.33 (m, 3H), 1.95 (m, 1H),
2.45–2.49 (m, 1H), 3.52–3.55 (m, 1H), 3.56–3.69 (m, 1H), 4.43 (m, 2H),
4.73 (dd, J = 24, 6.0 Hz, 1H), 5.00 (m, 1H), 5.42 (m, 1H), 5.60 (d, J =
19.6 Hz, 1H), 6.14 (d, J = 5.2 Hz, 1H), 11.25 (s, 1H). HRMS (ESI) Calcd
for C13H19N2O6, 299.12431; Found, 299.12320. 24R: 1H NMR (Methanol-d4,
500 MHz) δ 1.24 (m, 3H), 1.38 (m, 3H), 2.03–2.08 (m, 1H), 2.42–2.47 (m,
1H), 3.57–3.72 (m, 2H), 4.35 (m, 1H), 4.42 (dt, J1 = 21, 1.0 Hz, 1H), 4.51
(d, J = 19.6 Hz, 1H), 5.01 (m, 1H), 5.75 (d, J = 25.5 Hz, 1H), 6.30 (d,
J = 5.0 Hz, 1H). HRMS (ESI) Calcd for C13H19N2O6, 299.12431; Found,
299.12434.

Synthesis and Properties of DNA Containing Cyclonucleosides
677
3ꢀ-(2-Cyanoethyl N,N-diisopropyl phosphorodiamidite)-5ꢀ-(1-ethoxyethyl)-
6,5ꢀ(S)-cyclo-2ꢀ-deoxyuridine (25S) and 3ꢀ-(2-Cyanoethyl N,N-diisopropyl
phosphorodiamidite)-5ꢀ-(1-ethoxyethyl)-6,5ꢀ(R)-cyclo-2ꢀ-deoxyuridine (25R)
24S (184 mg, 0.62 mmol) was co-evaporated with acetonitrile (5 mL)
three times, and dissolved in acetonitrile (6.5 mL). The stirring reaction mix-
ture was cooled to 0◦C and 2-Cyanoethyl N,N,Nꢁ,Nꢁ-tetraisopropyl phospho-
rodiamidite (245 µL, 0.74 mmol), was added followed by tetrazole (0.45 M
in THF, 830 µL, 0.37 mmol). The reaction was allowed to warm at ambient
temperature and stirred for 4 hours before the solvent was removed under
reduced pressure. The crude residue was subjected to flash chromatogra-
phy (silica gel pre-washed by three column volumes of Hexanes/Et3N 99:1
followed by three column volumes of 100% Hexanes. Column eluent was
Ethyl Acetate/Hexane 60:40). This crude material dissolved in dry DCM
and precipitated from solution upon slow addition of hexanes to yield 25S
as a white foam (185.5 mg, 60.2%), Rf = 0.51 (Ethyl Acetate/Hexanes 75:25).
Compound 25R was obtained by the same procedure and isolated as a white
foam Rf = 0.49 (Ethyl Acetate/Hexanes 75:25). Compound 25S 31P NMR
(CDCl3, 200 M Hz) δ 146.5, 146.5, 146.2; HRMS (ESI-TOF) Calc. Mass for
C22H36N4O7P+ [M+H]+ 499.23216, found: 499.23142. Compound 25R 31P
NMR (CDCl3, 200 M Hz) δ 149.9, 149.5 149.2, 149.0; HRMS (ESI-TOF) Calc.
Mass for C22H36N4O7P+ [M+H]+ 499.23216, found: 499.23101.
Enzymatic Digestion
A mixture of 3 µL of alkaline phosphatase, 3 µL of phosphodiesterase I, 4
µL of nuclease P1, 10 µL of 10X alkaline phoaphatase buffer, 10 µL of 1.0 M
MgCl2, and 70 µL of water was added to a eppendorf tube containing 2 mmol
of the DNA oligomer. The mixture was left at 37◦C overnight, then cooled to
room temperature. The digested nucleoside mixture was analyzed by reverse-
phase analytical HPLC with a self-packed C18 column [0.46 cm 25 cm, 5 µm
˚
100 A, resins from Varian Analytical Instrument (Walnut Creek, CA, USA)].
Mobile phase buffer A: 100 mM triethylammonium acetate buffer (pH 7.0).
Mobile phase buffer B: acetonitrile. Flow rate: 1 mL/min. Gradient of the
elution: 0–25% in 30 minutes and 25%–100% in 10 minutes. The mixtures
were monitored by UV at 260 nm.
Thermal Denaturation Studies
All UV melting experiments were carried out in 20 mM pH 7.0 sodium
phosphate buffer with 100 mM or 1 M sodium chloride using AVIV Biomed-
ical Inc. (Lakewood, NJ, USA) spectrophotometer model 14DS UV-Vis with
a temperature controller. Self-complimentary sequence had a concentra-
tion of 1 µM while non-self-complementary sequences had concentrations
of 0.5 µM for each strand. The samples were heated to 95◦C for 5 minutes,

678
H. Yueh et al.
cooled to room temperature briefly, then further cooled to 4◦C. Samples
were placed in a 1 cm path length quartz cell with a stopper. The denatura-
tion experiments were carried out between 4◦C and 95◦C. Absorption data
was collected every degree at 260 nm. The Tm values were determined by
the first derivative of the melting profile using Origin and Microsoft Excel
software.
REFERENCES
1. Halliwell, B.; Gutteridge, J.M.C. Free Radicals in Biology and Medicine, Oxford University Press, Oxford,
New York, 2007.
2. Friedberg, E.C. DNA Repair and Mutagenesis, ASM Press, Washington, DC, 2006.
3. Sonntag, C.V. Free-Radical-Induced DNA Damage and its Repair: A Chemical Perspective, Springer, Berlin,
2006.
4. Dizdaroglu, M. Oxidative damage to DNA in mammalian chromatin. Mutat. Res. 1992, 275, 331–342.
5. Breen, A.P.; Murphy, J.A. Reactions of oxyl radicals with DNA. Free Radical Bio. Med. 1995, 18,
1033–1077.
6. Evans, M.D.; Dizdaroglu, M.; Cooke, M.S. Oxidative DNA damage and disease: Induction, repair and
significance. Mutat. Res.-Rev. Mutat. 2004, 567, 1–61.
7. Jaruga, P.; Dizdaroglu, M. 8,5ꢁ-Cyclopurine-2ꢁ-deoxynucleosides in DNA: Mechanisms of formation,
measurement, repair and biological effects. DNA Repair 2008, 7, 1413–1425.
8. Marietta, C.; Gulam, H.; Brooks, P.J. A single 8,5ꢁ-cyclo-2ꢁ-deoxyadenosine lesion in a TATA box
prevents binding of the TATA binding protein and strongly reduces transcription in vivo. DNA
Repair 2002, 1, 967–975.
9. Pullman, B.; Pullman, A. Submolecular structure of nucleic acids. Nature 1961, 189, 725–727.
10. Haromy, T.P.; Raleigh, J.; Sundaralingam, M. Enzyme-bound conformations of nucleotide substrates-
X-ray structure and absolute-configuration of 8,5ꢁ-cycloadenosine monohydrate. Biochemistry 1980,
19, 1718–1722.
11. Birnbaum, G.I.; Cygler, M.; Dudycz, L.; Stolarski, R.; Shugar, D. Comparison of solid-state and
solution conformations of R-Epimer and S-Epimer of 8,5ꢁ-cycloadenosine and their relevance to
some enzymatic-reactions. Biochemistry 1981, 20, 3294–3301.
12. Miaskiewicz, K.; Miller, J.H.; Fuciarelli, A.F. Theoretical-analysis of DNA intrastrand cross-linking by
formation of 8,5ꢁ-cyclodeoxyadenosine. Nucleic Acids Res. 1995, 23, 515–521.
13. Theruvathu, J.A.; Jaruga, P.; Dizdaroglu, M.; Brooks, P.J. The oxidatively induced DNA lesions
8,5ꢁ-cyclo-2ꢁ-deoxyadenosine and 8-hydroxy-2ꢁ-deoxyadenosine are strongly resistant to acid-induced
hydrolysis of the glycosidic bond. Mech. Ageing Dev. 2007, 128, 494–502.
14. Dirksen, M.L.; Blakely, W.F.; Holwitt, E.; Dizdaroglu, M. Effect of DNA conformation on the hydroxyl
radical-induced formation of 8,5ꢁ-cyclopurine 2ꢁ-deoxyribonucleoside residues in DNA. Int. J. Radiat.
Biol. 1988, 54, 195–204.
15. Brooks, P.J.; Wise, D.S.; Berry, D.A.; Kosmoski, J.V.; Smerdon, M.J.; Somers, R.L.; Mackie, H.;
Spoonde, A.Y.; Ackerman, E.J.; Coleman, K.; Tarone, R.E.; Robbins, J.H. The oxidative DNA lesion
8,5ꢁ-(S)-cyclo-2ꢁ-deoxyadenosine is repaired by the nucleotide excision repair pathway and blocks
gene expression in mammalian cells. J. Biol. Chem. 2000, 275, 22355–22362.
16. Kuraoka, I.; Bender, C.; Romieu, A.; Cadet, J.; Wood, R.D.; Lindahl, T. Removal of oxygen free-radical-
induced 5ꢁ,8-purine cyclodeoxynucleosides from DNA by the nucleotide excision-repair pathway in
human cells. P. Natl. Acad. Sci. USA 2000, 97, 3832–3837.
17. Juo, Z.S.; Chiu, T.K.; Leiberman, P.M.; Baikalov, I.; Berk, A.J.; Dickerson, R.E. How proteins recognize
the TATA box. J. Mol. Biol. 1996, 261, 239–254.
18. Suzuki, Y.; Matsuda, A.; Ueda, T. Synthesis of 6,5ꢁ-cyclo-2ꢁ,5ꢁ-dideoxypyrimidine nucleosides. Chem.
Pharm. Bull. 1987, 35, 1085–1092.
19. Theile, C.S.; McLaughlin, L.W. An efficient synthetic approach to 6,5ꢁ-(S)- and 6,5ꢁ-(R)-cyclouridine.
Chem. Commun. 2012, 48, 5587–5589.

Synthesis and Properties of DNA Containing Cyclonucleosides
679
20. Drew, H.R.; Wing, R.M.; Takano, T.; Broka, C.; Tanaka, S.; Itakura, K.; Dickerson, R.E. Structure of
a B-DNA dodecamer - Conformation and dynamics. P. Natl. Acad. Sci. USA-Biological Sciences 1981, 78,
2179–2183.
21. Partridge, B.L.; Salisbury, S.A. Structural studies on nucleic acids. NDB ID: BDL075, 1996.
22. Romieu, A.; Gasparutto, D.; Molko, D.; Cadet, J. Site-specific introduction of (5ꢁ S)-5ꢁ,8-cyclo-2ꢁ-
deoxyadenosine into oligodeoxyribonucleotides. J. Org. Chem. 1998, 63, 5245–5249.