Synthesis of the four stereoisomers of 2,3-epoxy-4-hydroxynonanal and their reactivity with deoxyguanosine

Article abstract of DOI:10.1039/c0ob00546k

2,3-Epoxy-4-hydroxynonanal (EHN) is a potential product of lipid peroxidation that gives rise to genotoxic etheno adducts. We have synthesized all four stereoisomers of EHN and individually reacted them with 2′-deoxyguanosine. In addition to 1,N²-etheno-2′- deoxyguanosine, 12 stereoisomeric products were isolated and characterized by ¹H NMR and circular dichroism spectroscopy. The stereochemical assignments were consistent with selective NOE spectra, vicinal coupling constants, and molecular mechanics calculations. Reversed-phase HPLC conditions were developed that could separate most of the adduct mixture.

Full text of DOI:10.1039/c0ob00546k

View Article Online / Journal Homepage / Table of Contents for this issue
Organic &
Biomolecular
Chemistry
Dynamic Article Links
Cite this: Org. Biomol. Chem., 2011, 9, 1960
www.rsc.org/obc
PAPER
Synthesis of the four stereoisomers of 2,3-epoxy-4-hydroxynonanal and their
reactivity with deoxyguanosine†
Katya V. Petrova, Donald F. Stec, Markus Voehler and Carmelo J. Rizzo*
Received 6th August 2010, Accepted 30th November 2010
DOI: 10.1039/c0ob00546k
2,3-Epoxy-4-hydroxynonanal (EHN) is a potential product of lipid peroxidation that gives rise to
genotoxic etheno adducts. We have synthesized all four stereoisomers of EHN and individually reacted
them with 2¢-deoxyguanosine. In addition to 1,N2-etheno-2¢-deoxyguanosine, 12 stereoisomeric
products were isolated and characterized by 1H NMR and circular dichroism spectroscopy. The
stereochemical assignments were consistent with selective NOE spectra, vicinal coupling constants, and
molecular mechanics calculations. Reversed-phase HPLC conditions were developed that could
separate most of the adduct mixture.
Introduction
Reactive oxygen and nitrogen species formed during normal
cellular respiration can initiate oxidative damage to biological
membranes through an oxygen-dependent, free radical chain
mechanism.1 The peroxidation of lipids gives a complex array
of products including simple enals (acrolein, crotonaldehyde and
higher congeners), more oxygenated enals (4-hydroxy-2-nonenal,
4-oxo-2-nonenal and malondialdehyde), and epoxyaldehydes (4,5-
epoxy-2-decenal) among other products.2 Lipid peroxidation has
been implicated as a contributing factor of human diseases such
as diabetes, cardiovascular injury, neurodegenerative diseases, and
cancer.3–7
4-Hydroxy-2-nonenal (HNE) is a major lipid peroxidation
product from w-6 polyunsaturated fatty acids.2 HNE is highly
cytotoxic and was shown to induce apoptosis in tumor cell lines.8,9
HNE reacts predominantly with Gua bases in DNA and has
mutagenic potential. Since HNE is produced as a racemate (1 and
2, Scheme 1), its reaction with 2¢-deoxyguanosine (dGuo) leads
to four diastereomeric adducts.10–12 We have previously reported
the synthesis of the oligonucleotides containing stereochemically
defined HNE-dGuo adducts and found that the adduct stereo-
chemistry can play a critical role in the chemistry, properties, and
mutagenicity of HNE-dGuo adducts.13,14
Model studies have shown that HNE can undergo epoxidation
with t-butylhydroperoxide to yield 2,3-epoxy-4-hydroxynonanal
Scheme
1
Stereoisomers of 4-Hydroxynonenal (HNE) and 2,3-e-
poxy-4-hydroxynonanal (EHN).
Departments of Chemistry and Biochemistry, Center in Molecular Toxicol-
ogy, and Vanderbilt-Ingram Cancer Center, Vanderbilt University, VU Sta-
tion B 351822, Nashville, Tennessee, 37235-1822, USA. E-mail: c.j.rizzo@
vanderbilt.edu; Fax: 1-615-343-1234; Tel: 1-615-322-6100
† Electronic supplementary information (ESI) available: Experimental
procedures for the synthesis of (4S)-HNE, selective 1H NOE spectra (17–
22), 1H homonuclear decoupling spectra (15, 16, 21, 22), CD spectra (11–
14, 16, 18–22), 1H (3, 5, 11–28) and 13C NMR spectra (3, 5, 11–14,23–28),
GC chromatograms (3, 5, 23–28), and HPLC chromatograms (11–22). See
DOI: 10.1039/c0ob00546k
(EHN),15–18 which is a more potent mutagen.19 It has been pro-
posed that lipid hydroperoxides may serve as the in vivo epoxidizing
agents, although the reaction of HNE with t-butylhydroperoxide
proceeds in modest yield.17,20 Blair has shown that 4,5-epoxy-
2-decenal is produced from the peroxidation of linolenic acid
suggesting that epoxyaldehydes can be direct products of lipid
peroxidation rather than a secondary product.21 Each enantiomer
1960 | Org. Biomol. Chem., 2011, 9, 1960–1971
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
of HNE can lead to a pair of diastereomeric epoxides; thus, four
stereoisomers of EHN are possible (3–6, Scheme 1).
characterized. In addition, reversed-phase HPLC conditions have
been developed that can separate most of these adducts.
Scheme 2 outlines the mechanism by which the EHN can
react with dGuo to give three types of DNA adducts, which are
the parent unsubstituted 1,N2-e-dGuo, C7-substituted e-dGuo
adducts, and tetracyclic adducts. EHN is a bis-electrophile, which
can react with dGuo at two separate nucleophilic sites to give
the 1,N2-cyclic carbinolamine adduct 8. The opposite sense of
the epoxide-opening, i.e., to give the 1,N2-dihydroxypropano-
dGuo adduct, has been observed as a minor product for the
reaction of dGuo with glycidaldehyde.22 However, products from
this competing pathway have not been observed for C3-substituted
2,3-epoxyaldehyde. The carbinolamine 8 can undergo reversible
dehydration to give imine 9, which is a common intermediate to
the three observed product types. Although an alternative pathway
for the loss of the side chain has been proposed previously,23
we favor the mechanism outlined by Golding involving loss of
the C7-sidechain as the aldehyde via a retro-aldol reaction (path
a) to give the unsubstituted 1,N2-etheno-dGuo (e-dGuo) adduct
10.22,24 Imine 9 can also undergo tautomerization to yield four
diastereomeric etheno adducts (11–14) possessing an intact C7-
sidechain (path b). Alternatively, imine 9 can be trapped by the
sidechain hydroxyl group to afford eight diastereomeric tetracyclic
adducts (15–22, path c). Thus, the reaction of racemic EHN with
DNA is expected to give a complex mixture of related DNA
adducts. Treatment of the C7-substituted e-dGuo or tetracyclic
adducts with base leads to a rapid conversion to the unsubstituted
1,N2-e-dGuo adduct 10. Related etheno adducts have also been
characterized from the reaction of EHN and dAdo.16,25 These
adducts have been observed from the treatment of DNA and cells
with a combination of HNE and t-butylhydroperoxide.17,25
Results and discussion
Asymmetric synthesis of all four stereoisomers of EHN
We have previously reported the synthesis of (4S)- and (4R)-trans-
HNE (1 and 2).9 The synthesis was based on the work of Yu and
Wang and proceeded in four steps from commercially available
2E-octenal.9,26 The absolute stereochemistry was established by
a Sharpless asymmetric epoxidation of 2E-octen-1-ol.13,26,27 The
HNE enantiomers were estimated to be >90% ee based on
optical rotation.26,28,29 Starting from (4S)- and (4R)-HNE, all
four diastereomers of 2,3-epoxy-4-hydroxynonanal (3–6) were
synthesized according to Scheme 3.
Scheme 2 Mechanism for the formation of the EHN adducts of
2¢-deoxyguanosine.
Scheme 3 Synthesis of EHN from HNE. Reagents: a) TES-Cl, DMAP,
Et3N (67–68%); b) NaBH4, THF (58–62%); c) Ti(OiPr)4, (+)-L-DET,
C6H5C(CH3)2OOH, CH2Cl2, -25 ◦C (73–77%); d) PhI(OAc)2, TEMPO
(64–76%); e) nBu4N+F-, THF (72–92%); f) Ti(OiPr)4, (-)-D-DET,
C6H5C(CH3)2OOH, CH2Cl2, -25 ◦C (74–75%).
We report here the synthesis of all four stereoisomers of
EHN. The individual isomers were then reacted with dGuo and
all four C7-substituted e-dGuo and eight tetracyclic adducts were
This journal is
The Royal Society of Chemistry 2011
Org. Biomol. Chem., 2011, 9, 1960–1971 | 1961
©

View Article Online
The hydroxyl group of (4S)-HNE (1) was protected as a tri-
ethylsilyl (TES) ether, then the aldehyde was reduced with NaBH4
in THF to give allyl alcohol 24 in ~40% overall yield. Sharpless
asymmetric epoxidation of 24 using the (+)-diethyltartrate ligand
gave the epoxyalcohol 25 with the syn relative stereochemistry
between the epoxide and the protected C4-hydroxyl group. The
primary alcohol was oxidized to aldehyde 27;30 NMR analysis of
27 showed two aldehyde resonances in a ~93 : 7 ratio, reflective
of the diastereoselectivity of the Sharpless epoxidation. Removal
of the silyl ether protecting group with fluoride gave (2S,3R,4S)-
EHN (3). When the (-)-diethyl tartrate ligand was used for the
Sharpless epoxidation of 24, the epoxyalcohol 26 possessing the
anti relative stereochemistry was obtained. Oxidation and depro-
tection afforded (2R,3S,4S)-EHN (5). The diastereoselectivity of
the Sharpless epoxidation of 26 was ~92 : 8 based on NMR analysis
of aldehyde 28. The remaining two stereoisomers of EHN, 4 and 6,
were synthesized in an identical fashion starting from (4R)-HNE
(2). The sequence required five steps from (4R)- or (4S)-HNE.
(16 and 18) face of the bicyclo[3.3.0]-ring system. Adducts 16
and 18, in which both substituents are on the more congested
concave face, are expected to be less favored than 15 and 17,
which place the substituents on the convex face. The products
from the cyclization of the anti diols (13 and 14) have one of
these substituents on the concave face and the other on the convex
face (19–22). The pure tetracyclic adducts would undergo ring
opening to the corresponding 7-substituted-e-dGuo adduct and
loss of the sidechain to 1,N2-e-dGuo over time; 16 and 18 were
particularly prone to this process, which complicated the 1H NMR
characterization as can be seen in the spectrum in Fig. 1 (right).
NMR analysis of the tetracyclic adducts 15–22
Comparison of the selective NOE spectra (600 MHz) when
protons H5a, H7, H8, and H8a were individually saturated for
all eight tetracyclic adducts 15–22 revealed patterns consistent
with the relative stereochemical assignments. The 1H NMR
assignments were made based on their COSY spectra and are
consistent with previous assignments.15,16 NOE enhancements for
vicinal protons (H5a–H8a, H7–H8, H8–H8a) about the furan ring
were in the range of 1.5–2.2% when syn and ~0.6–0.9% when anti.
The NOE enhancements for H5a–H7, H5a–H8, and H7–H8a were
between 0.1–0.7% when syn and <0.1% when anti. Although these
enhancements were small, they were reproducible and consistent
between samples. The NOE enhancements are summarized in
Table 1 and an example of the selective NOE spectra of adducts
15 and 16 are shown in Fig. 1.
Reaction of EHN with dGuo
The reaction of EHN with dGuo was reported to proceed slowly
and in low yield.15 The reactivity can be increased at higher pH;
however, a greater proportion of the 1,N2-e-dGuo was observed
under such conditions. We found that the individual stereoisomers
of EHN (3–6) react with dGuo in DMF and K2CO3 (12–17 h) to
give a single stereoisomer of the 7-substituted etheno adducts (11–
14) in good yield (62–71%) and only a small amount of 1,N2-e-
dGuo (10) was observed (Scheme 4).31 Interestingly, the tetracyclic
adducts (15–22) were not observed under these conditions. It
should be noted that the stereochemistry of the allylic hydroxyl
group of the C7-sidechain (C10) of 11–14 can slowly scramble
over time.31
The conformations of the four relative stereoisomers of the
tetracyclic adducts were determined by molecular mechanics using
the MMX force field as implemented in PC Model (v. 9.2, Serena
Software). The modified Gua bases of 15/17, 16/18, 19/21, and
20/22 have the same relative stereochemistry. Methyl and ethyl
groups were used in place of the deoxyribose unit and C7-pentyl
chain, respectively, to simply the analysis. Molecular mechanics
were performed on adducts 15¢, 16¢, 19¢, and 20¢, where the
prime (¢) notes the simplified model compound. The predicted
vicinal H5a–H8a, H7–H8, and H8–H8a coupling constants were
computed from the dihedral angles and compared to experimental
Formation of the tetracyclic adducts
Each of the 7-substitued-e-dGuo adducts 11–14 were individually
incubated in 25 mM, pH 6.8 phosphate buffer. Each was consumed
over 2–4 days to give two new products, which were identified as the
tetracyclic adducts (15–22, Scheme 4). The cyclization reaction was
performed under high dilution conditions to prevent dimerization
of the 7-substituted-e-dGuo. The dimers were not characterized,
but this processes has been reported for other 7-(1-hydroxyalkyl)-
e-dGuo adducts.22,31,32 The mechanism of the cyclization involves
tautomerization back to the imine (9) followed by cyclization
(Scheme 2). The tautomerization can generate two diastereomers
at C7, which corresponds to the C8a-position of the tetracyclic
product. Bicyclo[3.3.0]octane systems have a strong preference
for a cis-ring fusion;33,34 thus, each diastereomer of the imine is
predicted to cyclize to a specific tetracyclic adduct. In the case
of 13 and 14, where the C7-sidechain vicinal diols possess an
anti relative stereochemistry, the cyclization proceeds to afford
near equal amounts of the two tetracyclic adducts. The cyclization
was more selective in the case of the syn vic-diol (11 and 12),
affording a major and minor product in ~7 : 1 ratio. The relative
stereochemistry of the tetracyclic adducts (15–22) explains the
observed product ratio. Cyclization of the syn diols (11 and 12)
is predicted to give a product in which the C7-pentyl and C8-
hydroxyl groups are either on the convex (15 and 17) or concave
1
values (Table 2) obtained from a series of 1D H homonuclear
decoupling experiments (Fig. S7–S10, ESI†).35 The rigidity of the
bicyclo[3.3.0]octane fused to the Gua base makes this system ideal
for this analysis and there is reasonable agreement between the
predicted and observed values for three of the four adducts. The
bridgehead H5a and H8a protons are syn for all the compounds
and the dihedral angle between them is predicted to be small
Table 1 Percentage NOE enhancement when the first proton listed for
each pair is saturated (NOE enhancement when the second proton is
saturated)
H5a–H7 H5a–H8 H5a–H8a H7–H8
H7–H8a H8–H8a
15 0.1 (0.1) 0.1 (0.1) 1.8 (1.8)
16 0.7 (0.7) 0 (0) 1.9 (2.0)
17 0.1 (0.2) 0.1 (0.1) 1.8 (1.8)
18 0.7 (0.7) 0 (0) 1.5 (1.9)
19 0 (0) 0.1 (0.1) 1.7 (1.9)
20 0.4 (0.4) 0.1 (0.1) 1.7 (1.7)
21 0.1 (0.1) 0.3 (0.2) 2.0 (2.0)
22 0.4 (0.4) 0.1 (0.1) 1.6 (1.5)
1.8 (1.8) 0.1 (0.1) 0.9 (0.8)
1.8 (1.8) 0.6 (0.6) 2.2 (2.0)
1.7 (1.7) 0.1 (0.1) 0.8 (0.8)
1.7 (1.8) 0.6 (0.6) 2.1 (2.0)
0.6 (0.6) 0 (0)
1.9 (1.9)
0.7 (0.8) 0.2 (0.2) 0.8 (0.7)
0.6 (0.8) 0.1 (0.1) 1.9 (2.0)
0.7 (0.7) 0.1 (0.1) 0.7 (0.7)
1962 | Org. Biomol. Chem., 2011, 9, 1960–1971
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
Scheme 4 2¢-Deoxyguanosine adducts of EHN.
Fig. 1 Selective NOE spectra of tetracyclic adducts 15 and 16. The NOE spectra for the remaining adducts can be found in Fig. S1–S6, ESI.†
This journal is The Royal Society of Chemistry 2011 Org. Biomol. Chem., 2011, 9, 1960–1971 | 1963
©

View Article Online
Table 2 The observed vicinal 1H–1H coupling constants for tetracyclic
adducts 15–22 and the calculated values from MMX minimized structures
Observed
Calculated (dihedral angle)
J5a–8a
6.6
7.7
6.6
7.7
6.9
6.7
6.9
6.7
J7–8
2.7
2.6
2.7
2.6
8.0
2.6
8.0
2.5
J8–8a
0.3
5.8
0.3
5.8
7.1
1.2
7.1
1.2
J5a–8a
J7–8
J8–8a
15
16
17
18
19
20
21
22
7.2 (1.8◦)
7.2 (8.1◦)
1.8 (47◦)
2.2 (44◦)
0.7 (89◦)
7.5 (19◦)
7.5 (4.1◦)
7.3 (6.5◦)
8.2 (161◦)
8.3 (163◦)
8.1 (22◦)
5.4 (140◦)
(1.8◦–4.0◦) with J values between 7.2 and 7.6 Hz; the observed val-
ues were between 6.6 and 7.7 Hz. The terminal furan is predicted
to adopt an envelope conformation for all four stereoisomers in
which C5a, O6, C8, and C8a were nearly coplanar. The C7 atom
is predicted to occupy an endo position relative to the other rings
for 15¢ and 19¢, whereas the C7 atom is predicted to be exo for
16¢ and 20¢; this designation can be seen for the MMX minimized
structures in Fig. 2.
The H7 and H8 protons are syn for 15¢ and 16¢ and are predicted
to have dihedral angles of 47◦ and 44◦, corresponding to small
coupling constants of 1.8 and 2.2 Hz, respectively. These vicinal
coupling constants are in good agreement with the observed values
of 2.7 and 2.6 Hz. The relative configuration of H8 and the
bridgehead H8a protons is anti for 15¢ and syn for 16¢. The H8–
H8a dihedral angles are calculated to be 89◦ and 19◦ with predicted
coupling constants of 0.7 and 7.5 Hz, respectively. Once again these
values are in reasonable agreement with the observed values of 0.3
and 5.8 Hz. A correlation was not observed between H8 and H8a
in COSY spectra of 15/17, indicating that the dihedral angle was
~90◦ for these protons and is consistent with the calculated value.
The relative stereochemistry of the H7 and H8 is anti in 19¢
and 20¢. The H7 and H8 protons of 19¢ are situated on the exo
and endo faces of the ring system, respectively. The calculated H7–
H8 dihedral angle is 161◦ leading to a predicted vicinal coupling
constant of 8.2 Hz; this value is in excellent agreement with the
observed value of 8.0 Hz. The relative stereochemistry of H8 and
H8a protons is syn and predicted to have a dihedral angle of 22◦.
The calculated coupling constant for this angle is 8.1 Hz, which is
again in good agreement with the observed value of 7.1 Hz.
The H7 and H8 protons of the MMX minimized conformation
of 20¢ occupy pseudo diaxial dispositions off the furan ring, similar
to 19¢, resulting in a large predicted dihedral angle of 163◦. The
H8 and H8a protons are anti with a dihedral angle of 140◦. The
calculated H7–H8 and H8–H8a vicinal coupling constants are
8.3 and 5.4 Hz, respectively, which is in poor agreement with
the observed values of 2.6 and 1.2 Hz. A conformational search
(GMMX), in which the bonds of the furan ring were varied,
was performed as implemented in PC Model using the default
parameters. A conformation 5.9 kJ mol-1 (1.4 kcals mol-1) higher
in energy was found that better matched the observed coupling
constants. The main difference between the two conformations is
the disposition of C7. The C7 atom is exo in the lowest energy
conformation (20¢) and endo in the higher energy conformation
(20¢¢). The C7-endo conformation (20¢¢) places H7 and H8 in
pseudo equatorial positions with a dihedral angle of 89◦, while
the H8–H8a dihedral angle is predicted to be 95◦. These dihedral
Fig. 2 MMX minimized conformation of the four relative stereoisomers
of the tetracyclic adducts.
angles are predicted to give rise to small coupling constants (0.9
and 0.6 Hz, respectively), which are in better agreement with the
observed values.
1964 | Org. Biomol. Chem., 2011, 9, 1960–1971
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
The C7-endo conformation of 20¢¢ also agrees with the observed
NOE enhancements as compared to the other tetracyclic adduct.
The H5a–H7 and H8a–H7 distances when H7 is in the pseudo-
axial position for the C7-exo conformation of 20¢ are 2.7 and 3.0
was observed in the CD spectra of 7-substituted etheno adduct
11 and 12, as well as 13 and 14. In the case of adducts 11–14,
the sign of the CD absorbance at ~295 nm was correlated to
the C1-stereochemistry of the C7-(1,2-dihydroxyheptyl) sidechain
(see Fig. S14 and S15, ESI†). The 1S-stereochemistry resulted
in a positive ellipticity (11 and 14), while the 1R-stereochemistry
resulted in a negative ellipticity (12 and 13).
˚
A, respectively. These values are nearly identical to those of 16¢
˚
(2.7 and 3.1 A) in which H7 is also predicted to be pseudo-axial.
The H5a–H7 and H8a–H7 NOE enhancements for 16 were 0.7
and 0.6%, respectively, as compared to 0.4 and 0.2% for 20. The
H5a–H7 and H8a–H7 distances predicted when H7 is pseudo-
equatorial as for the C7-endo conformation of 20¢¢ are 3.8 and
HPLC separation of the EHN-dGuo adducts
The reaction of EHN with dGuo afforded at least 13 prod-
ucts, including 1,N2-e-dGuo (10), four diastereoisomers of 7-
(1,2-dihydroxyheptyl)-e-dGuo (11–14), and eight diastereomeric
tetracyclic adducts (15–22). Unfortunately, we were unable to
develop HPLC conditions that could resolve all 13 products.
Optimal conditions were developed in which 11 peaks could
be observed and their identities were verified by comparison of
the retention time to the authentic compound (Fig. 4). C7-(1,2-
Dihydroxyheptyl)-e-dGuo adducts 11 and 12 coeluted as did the
C7-substituted e-dGuo adduct 13 and tetracyclic adduct 20; in
addition, tetracyclic adducts 16 and 18 were not well resolved, but
distinguishable.
˚
4.1 A; these longer distances are more consistent with the lower
NOE enhancements as compared to 16. Taken together, the vicinal
coupling constants and NOE data are more consistent with the C7-
endo conformation for adducts 20/22 over the energetically more
favorable C7-exo conformation. The reason for this preference is
not entirely clear. The NMR experiments were performed in 1 : 1
CD3CN/D2O; hydrophobic considerations would favor the more
compact C7-endo conformation, which is likely to have a smaller
surface area.
Circular dichroism spectra
The eight tetracyclic adducts 15–22 are diastereomers. However,
the modified guanine base of adducts 15 and 17, 16 and 18, 19
and 21, and 20 and 22 have enantiomeric relationships, which is
reflected in the sign of their CD spectra.13,15,36,37 An example is
shown in Fig. 3 for 15 and 17. The sign of the CD absorbance at
~275 nm was correlated to the stereochemistry at the bridgehead
positions, C5a and C8a. A positive ellipticity was observed for the
(5aR,8aS)-stereochemistry (15, 18, 19, and 22) and a negative
ellipticity was observed for the (5aS,8aR)-stereochemistry (16,
17, 20, and 21). CD spectra for the remaining adducts can be
found in Fig. S11–S13, ESI.† The same enantiomeric relationship
Fig. 4 Reversed-phase HPLC analysis of a mixture of all dGuo adducts
of EHN (10–22).
Conclusions
The potential role of EHN as an endogenous toxicant was hypoth-
esized by Sodum and Chung who observed EHN-dGuo adducts
when dGuo was reacted with HNE using THF as a co-solvent;18
it was proposed that the hydroperoxide of THF epoxidized
HNE. Other hydroperoxides such as H2O2, t-butylhydroperoxide,
and lipid hydroperoxides can also carry out the epoxidation
reaction. Subsequently, the reaction of partially separated, racemic
syn and anti diastereomers of EHN with dAdo and dGuo was
examined.15,16 Six products were characterized from the reaction
Fig. 3 CD spectra of adducts 15 and 17. The enantiomeric relationship
of the modified Gua bases gives rise to CD signals of opposite signs at
~275 nm.
This journal is
The Royal Society of Chemistry 2011
Org. Biomol. Chem., 2011, 9, 1960–1971 | 1965
©

View Article Online
with dGuo, including 1,N2-e-dGuo, a 7-(1,2-dihydroxyheptyl)-e-
dGuo, and four of the tetracyclic adducts, although the absolute
stereochemistry was not determined. These adducts were also
observed when calf thymus DNA was treated with EHN.16,17,25
The tetracyclic adducts were subject to acid deglycosylation
and the corresponding Gua bases had opposite CD spectra,
suggesting enantiomeric relationships of the modified bases. The 7-
(1,2-dihydroxyheptyl)-e-dGuo, and four tetracyclic adducts were
rapidly converted to 1,N2-e-dGuo under basic conditions. This
work established a possible mechanism for the endogenous for-
mation of promutagenic etheno lesions. Other lipid peroxidation
products have also been shown to give rise to 1,N2-e-dGuo.31,38–42
In the present work, we synthesized the four stereoisomers
of EHN starting from (4R)- and (4S)-HNE. The individual
EHN stereoisomers were reacted with dGuo to afford a single
diastereomer of 7-(1,2-dihydroxyheptyl)-e-dGuo with minimal
formation of 1,N2-e-dGuo. Each 7-(1,2-dihydroxyheptyl)-e-dGuo
isomer was then incubated in phosphate buffer to give two
tetracyclic adducts. The absolute stereochemistries of all adducts
were established. The relative stereochemistry of the tetracyclic
adducts was consistent with selective NOE enhancements and
the vicinial coupling constants were consistent with predicted
values that were calculated from dihedral angles as determined
by molecular mechanics.
(30 m, 0.25 mm ID, 0.25 mm). Helium was used as a carrier gas
with 1 mL min-1 flow rate. The temperature program ◦was follows:
◦
40 ◦C (6 min), 7 C min-1 to 70 ◦C and then 10 C min-1 to
230 ◦C (hold 13 min). Optical rotations were measured on an
Autopol IV polarimeter, using a 1.0 dm cell and CHCl3 as a
1
solvent. All H and 13C NMR spectra for the isomers of EHN
and their precursors were acquired on either a Bruker DRX-300
or AV-400 spectrometer in CDCl3 and CD2Cl2 which also served as
an internal chemical shift standard. Data are reported as follows:
chemical shifts, multiplicity, integration, coupling constant (Hz).
13C chemical shifts are reported in ppm with CD2Cl2 serving as
an internal reference at 53.31 ppm. ECD spectra were recorded
on Jasco J720 Spectropolarimeter. A Beckman gradient HPLC
system (32 Karat software version 7.0, pump module 125) with a
diode array UV detector (module 168) monitoring at 260 nm was
used for separations with YMC ODS-AQ columns (Waters Corp).
(4S)-4-Hydroxy-2E-nonenal (1). (4S)-HNE (1) was prepared
as previously described.9,26 [a]D24.5 +44.7 (c 0.36, CHCl3) (lit:29 [a]2D0 =
+48◦ (c 0.69, CHCl3)); dH (400 MHz, CD2Cl2) 0.91 (t, 3H, J =
6.9 Hz, CH3), 1.29–1.51 (m, 6H, 3CH2), 1.54–1.69 (m, 2H, CH2),
1.86 (br.s., 1H, OH), 4.43 (m, 1H, J = 5.2 Hz, H-4), 6.27 (ddd, 1H,
J = 1.6, 7.9, 15.7 Hz, H-2), 6.84 (dd, 1H, J = 4.6, 15.7 Hz, H-3),
9.57 (d 1H, J = 7.9 Hz, CHO); m/z (GC-EI) 157 (100, M++1), 139
(25), 109 (40), 81 (30).
(4R)-Hydroxy-2E-nonenal (2). (4R)-HNE (2) was prepared in
the same manner as 1. [a]2D5 -46.7 (c 0.73, CHCl3) (lit:29 [a]2D5 -46
(c 0.45, CHCl3)).
Experimental section
General
(4S)-4-Triethylsilyloxy-2E-nonen-1-al (23).
A solution of
All commercially obtained chemicals were used as received except
(methoxymethyl)triphenylphosphonium chloride, which was used
after drying overnight in an Abderhalden apparatus under vacuum
at 78 ◦C. CH2Cl2 was freshly distilled from calcium hydride. An-
hydrous THF was freshly distilled from a sodium/benzophenone
ketyl. All reactions were performed under an Ar atmosphere.
Glassware was flame-dried and cooled under Ar. Molecular sieves
(4S)-HNE (1, 0.52 g, 3.36 mmol), 4-(dimethylamino)pyridine
(0.049 g, 0.4 mmol) and triethylamine (1.02 g, 10 mmol) in dry
CH2Cl2 (25 mL) was cooled in an ice-bath. Chlorotriethylsilane
(0.60 g, 3.98 mmol) was added dropwise to the stirred solution; the
reaction mixture was stirred at 0◦ C for 10 min, then the ice bath
was removed and mixture stirred at room temperature until the
starting material was consumed (2 h). Water (25 mL) was added
with stirring. The organic layer was separated, and the aqueous
phase was extracted with CH2Cl2 (3 ¥ 25 mL). The combined
organic layers were dried over MgSO4, filtered, and concentrated
in vacuo. Purification by flash chromatography on silica, eluting
with 5% ether in hexanes, afforded 23 (0.62 g, 68%). [a]2D4 +17.8 (c
0.43, CHCl3). dH (400 MHz, CD2Cl2) 0.62 (m, 6H, 3CH2Si), 0.88
(t, 3H, J = 6.8 Hz, CH3), 0.95 (m, 9H, 3CH3CH2Si), 1.28–1.37
(m, 6H, 3CH2), 1.54–1.62 (m, 2H, CH2), 4.43 (tdd, 1H, J = 1.5,
4.6, 6.1 Hz, H4), 6.23 (ddd, 1H, J = 1.5, 8.0, 15.5 Hz, H2), 6.82
(dd, 1H, J = 4.8, 15.5 Hz, H3), 9.56 (d, 1H, J = 8.0 Hz, CHO);
dC (100 MHz, CD2Cl2) 193.9, 160.5, 130.9, 72.0, 37.7, 32.2, 24.9,
23.0, 14.2, 7.0 (3), 5.2 (3); m/z (GC-EI) 271 (7, M++1), 253 (100),
241 (50), 157 (28), 139 (12), 81(8).
˚
(4 A) were activated with a microwave oven for at least 7 min
and cooled in a desiccator. Flash column chromatography was
performed using silica gel (32–63 mm; 230 ¥ 450 mesh). Analytical
thin layer chromatography was performed on silica gel glass
plates (Merck, silica gel 60 F254 layer thickness 250 mm) and
visualized by staining with p-anisaldehyde followed by charring.
High-resolution FAB mass spectra for compounds 11–15, 17 were
obtained from the University of Notre Dame Mass Spectrometry
Center using nitrobenzyl alcohol (NBA) as a matrix. High-
resolution mass spectra for compounds 16, 18–22 were obtained
from the Vanderbilt University Mass Spectrometry Resource
Center and recorded on a Waters Synapt hybrid quadrupole oa-
TOF high resolution mass spectrometer operated in “w” mode,
posESI. A post-acquisition gain correction factor was applied
using sodium formate as lock mass. Qualitative GC-MS analyses
were performed on a Varian Saturn 2100T mass spectrometer
using the following conditions: electron-impact ionization (EI)
mass spectra were generated at 70 eV; ion source temperature
was 200 ◦C and electron multiplier voltage was 1400 V. Scanning
was performed from m/z 40–450. Samples were introduced via
Varian 3900 GC equipped with CP 8410 autoinjector using the
following conditions: split mode injection (ratio 100 : 50) at 180 ◦C
on high resolution gas chromatography column Agilent HP-5 MS
(4R)-4-Triethylsilyloxy-2E-nonenal (ent-23). ent-23 was pre-
pared from (4R)-HNE (2) in 67% yield following the procedure
described above for 23. [a]2D4.2 -16.0 (c 0.45, CHCl3).
(4S)-4-Triethylsilyloxy-2E-nonen-1-ol (24). Sodium borohy-
dride (0.11 g, 2.82 mmol) was added in portions to a stirred
solution of 23 (0.61 g, 2.25 mmol) in dry THF (10 mL) under Ar.
The reaction mixture was stirred for 30 min at room temperature,
and then deionized water (25 mL) was added. The resulting
1966 | Org. Biomol. Chem., 2011, 9, 1960–1971
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
afford 26 in 74% yield. [a]2D4.6 +11.3 (c 0.47, CHCl3); dH (300 MHz,
CD2Cl2) 0.57–0.65 (m, 6H, 3CH2Si), 0.88 (t, 3H, J = 6.7 Hz,
CH3), 0.92–0.98 (m, 9H, 3CH3CH2Si), 1.28–1.32 (m, 6H, 3CH2),
1.47–1.54 (m, 2H, CH2), 1.81 (br s, 1H, OH), 2.89 (dd, 1H, J =
2.3, 6.6 Hz, H3), 2.97 (quintet, 1H, J = 2.4 Hz, H2), 3.37 (q, 1H,
J = 6.5 Hz, H4), 3.56 (d, 1H, J = 4.6, 12.8 Hz, H1b), 3.91 (d, 1H,
J = 2.3, 12.6 Hz, H1a); dC (75 MHz, CD2Cl2) 74.0, 61.9, 59.6,
56.8, 35.3, 32.3, 24.3, 22.9, 14.1, 6.9 (3), 5.2 (3); m/z (GC-EI) 289
(0.5, M++1), 241 (100), 215 (60), 157 (40), 139 (80), 103 (60).
mixture was extracted with CH2Cl2 (3 ¥ 30 mL), the combined
organic layers were washed with water, dried over MgSO4, filtrated,
and evaporated. Purification by flash chromatography on silica,
eluting with 10% ether in hexanes, afforded 24 (0.38 g, 62%). [a]D24.2
-2.6 (c 0.38, CHCl3); dH (400 MHz, CD2Cl2) 0.55–0.62 (m, 6H,
3CH2Si), 0.89–0.99 (m, 12H, 3CH3CH2Si, -CH3), 1.24–1.4 (m, 6H,
3CH2), 1.54–1.55 (m, 2H, CH2), 1.59 (br d, 1H, OH), 4.09–4.13
(m, 3H, CH2-1, H4), 5.62–5.77 (m, 2H, H2, H3); dC (100 MHz,
CD2Cl2) 134.9, 128.6, 72.7, 62.9, 38.2, 31.8, 24.8, 22.6, 13.7, 6.5
(3), 4.8 (3); m/z (GC-EI) 271 (2), 255 (100), 243 (25), 201 (25), 123
(15).
(2R,3S,4R)-2,3-Epoxy-4-triethylsilyloxy-2E-nonan-1-ol (ent-
26). The Sharpless epoxidation of ent-24 was performed in the
same manner as described above for 25 using diethyl L-(+)-tartrate
to afford ent-26 in 77% yield. [a]2D3.9 -8.9 (c 0.27, CHCl3).
(4R)-4-Triethylsilyloxy-2E-nonen-1-ol (ent-24). ent-24 was
prepared in 58% yield from ent-23 following the procedure
described above for 24. [a]2D4.7 +2.5 (c 0.32, CHCl3).
(2S,3S,4S)-2,3-Epoxy-4-triethylsilyloxy-2E-nonan-1-al
(27).
(2R,3S,4S)-2,3-Epoxy-4-triethylsilyloxy-2E-nonan-1-ol (25).
A stirred suspension of activated powdered 4 A molecular
sieves (0.075 g) and anhydrous CH2Cl2 (15 mL) was cooled
under Ar to -40◦ C. In a separate flask, diethyl L-(+)-tartrate
(0.027 g, 0.13 mmol) was stirred in anhydrous CH2Cl2 (3 mL)
(Diacetoxyiodo)benzene (0.135 g, 0.42 mmol) was added to a
solution of 25 (0.11 g, 0.38 mmol) and 2,2,6,6-tetramethyl-1-
piperidinyloxyl (TEMPO, 0.006 g, 0.038 mmol) in 10 mL dry
CH2Cl2. The reaction mixture was stirred at room temperature
until 25 was no longer detectable by TLC, and then diluted
with CH2Cl2 (20 mL). The mixture was washed with a saturated
aqueous solution of Na2S2O3 (10 mL) and the aqueous layer
was extracted with CH2Cl2 (2 ¥ 10 mL). The combined organic
layers were washed with 5% NaHCO3 (10 mL) and brine,
dried over MgSO4, filtrated, and evaporated. Purification by flash
chromatography on silica, eluting with 3% ethyl acetate in hexanes
afforded 27 (0.07 g, 64%); dH (300 MHz, CD2Cl2); 0.55–0.63 (m,
6H, 3CH2Si), 0.87–0.97 (m, 12H, CH3, 3CH3CH2Si), 1.27–1.43
(m, 6H, 3CH2), 1.50–1.58 (m, 2H, CH2), 3.20 (dd, 1H, J = 1.9,
3.7 Hz, H3), 3.38 (dd, 1H, J = 1.8, 6.3 Hz, H2), 3.79–3.84 (m, 1H,
H4), 9.08 (d, 1H, J = 6.3 Hz, CHO); dC (75 MHz, CD2Cl2) 198.7,
69.7, 59.4, 56.7, 35.4, 32.2, 24.7, 22.9, 14.1, 6.8 (3), 5.1 (3); m/z
(GC-EI) 287 (7.5, M++1), 269 (75), 241 (100), 227 (50) 215 (30),
157 (65), 109 (30), 81 (20).
˚
˚
over activated 4 A molecular sieves (about 0.1 g) for 15 min, then
transferred to the reaction flask via syringe. In separate flasks,
titanium tetraisopropoxide (0.031 g, 0.11 mmol) and cumene
hydroperoxide (88%, 0.21 g, 1.38 mmol) were stirred in CH2Cl2 (5
˚
mL) over activated 4 A molecular sieves (about 0.2 g) for 15 min,
then consecutively added dropwise to the reaction mixture via
syringe. The reaction mixture was stirred for 40 min at -40◦
C
under Ar. A solution of 24 (0.15 g, 0.55 mmol) in CH2Cl2 (15
˚
mL) was stirred over activated 4 A molecular sieves (about 0.1
g) for 15 min, and then added to the reaction flask dropwise via
syringe. The reaction mixture was stirred at -25◦ C for 4 h (until
24 was no longer detectable TLC); the reaction was quenched
by the addition of a solution of ferrous sulfate (15.0 g) in 10%
aqueous tartaric acid (3 mL) to the reaction mixture at -25◦
C. The mixture was stirred for 1h, and then the organic phase was
separated, washed with water (2 ¥ 10 mL), dried over MgSO4,
filtrated, and concentrated. The residue was diluted with ether
(75 mL) and stirred with 30% NaOH in saturated brine (2 mL) at
0◦ C for 20 min. The organic layer was separated, washed with
brine, dried over MgSO4, filtrated and evaporated. Purification
by flash chromatography on silica, eluting with 5% ethyl acetate
in hexanes afforded 25 (0.115 g, 73%). [a]2D5 -15.7 (c 0.23, CHCl3);
dH (300 MHz, CD2Cl2) 0.55–0.64 (m, 6H, 3CH2Si), 0.90 (t, 3H,
J = 6.8 Hz, CH3), 0.93–0.99 (m, 9H, 3CH3CH2Si), 1.29–1.40 (m,
6H, 3CH2), 1.55–1.58 (m, 3H, OH, CH2), 2.93 (dd, 1H, J = 2.2,
4.7 Hz, H3), 3.13 (quintet, 1H, J = 2.2 Hz, H2), 3.61–3.66 (m,
2H, H1b, H4), 3.95 (dd, 1H, J = 2.0, 12.7, H1a); dC (75 MHz,
CD2Cl2) 71.1, 62.0, 58.3, 56.7, 35.6, 32.3, 24.8, 23.0, 14.2, 6.9 (3),
5.2 (3); m/z (GC-EI) 289 (0.25, M++1), 241 (100), 215 (65), 157
(50), 139 (80), 103 (75).
(2R,3R,4R)-2,3-Epoxy-4-triethylsilyloxy-2E-nonan-1-al (ent-
27). ent-27 was prepared from ent-25 in 76% yield following the
procedure described above for 27. [a]2D3.6 -33.5 (c 0.46, CHCl3).
(2R,3R,4S)-2,3-Epoxy-4-triethylsilyloxy-2E-nonan-1-al (28).
28 was prepared from 26 in 76% yield following the procedure
described above for 27. [a]2D4.7 -59.6 (c 0.46, CHCl3); dH NMR
(300 MHz, CDCl3); 0.58–0.68 (m, 6H, 3CH2Si), 0.89 (t, 3H, J =
6.8 Hz, CH3), 0.94–1.0 (m, 9H, 3CH3CH2Si), 1.22–1.42 (m, 6H,
3CH2), 1.50–1.60 (m, 2H, CH2), 3.22–3.26 (m, 2H, H3, H4),
3.44 (q, 1H, J = 6.2 Hz, H2), 9.02 (d, 1H, J = 6.3 Hz, CHO); dC
(75 MHz, CD2Cl2) 198.2, 73.0, 60.2, 57.6, 35.1, 32.1, 25.1, 22.8,
14.1, 6.9 (3), 5.1 (3); m/z (GC-EI) 287 (5, M++1), 269 (65), 241
(100), 227 (50) 215 (28), 157 (100), 109 (25), 81 (15).
(2S,3S,4R)-2,3-Epoxy-4-triethylsilyloxy-2E-nonan-1-al (ent-
28). ent-28 was prepared from ent-26 in 76% yield following the
procedure described above for 28.
(2S,3R,4R)-2,3-Epoxy-4-triethylsilyloxy-2E-nonan-1-ol (ent-
25). The Sharpless epoxidation of ent-24 was performed
in the same manner as described above for 25 using diethyl
D-(-)-tartrate to afford ent-25 in 75% yield. [a]2D3.6 +16.8 (c 0.45,
CHCl3).
(2S,3R,4S)-2,3-Epoxy-4-hydroxy-2E-nonan-1-al
(3).
Tetrabutylammonium fluoride (0.25 mL, 1.0 M in THF)
was added to a stirred solution of 27 (0.06 g, 0.21 mmol) in
THF (2 mL) and the reaction was stirred for 1 h at ambient
temperature. The solvent was then removed under reduced
pressure. Purification by flash chromatography on silica, eluting
with 2% methanol in CH2Cl2 afforded 3 (0.03 g, 83%). [a]2D3.1 +4.8
(2S,3R,4S)-2,3-Epoxy-4-triethylsilyloxy-2E-nonan-1-ol (26).
The Sharpless epoxidation of 24 was performed in the same
manner as described above for 25 using diethyl D-(-)-tartrate to
This journal is
The Royal Society of Chemistry 2011
Org. Biomol. Chem., 2011, 9, 1960–1971 | 1967
©

View Article Online
(c 0.84, CH2Cl2); dH (400 MHz, CD2Cl2); 0.91 (t, 3H, J = 6.8 Hz,
CH3), 1.28–1.62 (m, 8H, 4CH2), 1.77 (br s, 1H, OH), 3.31 (dd, 1H,
J = 2.1, 2.8 Hz, H3), 3.45 (dd, 1H, J = 1.8, 6.0 Hz, H2), 3.96–3.99
(m, 1H, J = 3.3 Hz, H4), 9.14 (d, 1H, J = 6.1 Hz, CHO); dC
(75 MHz, CD2Cl2) 198.5, 68.3, 59.5, 55.9, 33.7, 32.1, 25.1, 22.9,
14.1; m/z (GC-EI) 173 (0.5, M++1), 83 (95), 71 (100), 55 (90).
H13b, H14, H15), 1.27–1.37 (m, 3H, H13a,H12), 2.34–2.38 (m,
1H, H2¢¢), 2.66–2.71 (m, 1H, H2¢), 3.63 (dd, 1H, J = 4.2, 12.3 Hz,
H5¢¢), 3.70 (dd, 1H, J = 3.6, 12.3 Hz, H5¢), 3.86–3.90 (m, 1H,
H11), 3.98–4.0 (m, 1H, H4¢), 4.48–4.51 (m, 1H, H3¢), 5.01 (d, 1H,
J = 5.6 Hz, H10), 6.30 (dd, 1H, J = 6.3, 7.4 Hz, H1¢), 7.17 (s,
1H, H6), 7.97 (s, 1H, H2); dC (150.9 MHz, D2O/CD3CN) 155.6,
150.4, 147.6, 139.8, 125.4, 117.3, 116.7, 88.3, 85.6, 74.2, 72.3, 70.4,
62.9, 39.8, 33.7, 32.2, 25.8, 23.0, 14.3: m/z (FAB) 422.2050 (MH+,
C19H28N5O6 requires 422.2040).
(2R,3S,4R)-2,3-Epoxy-4-hydroxy-2E-nonan-1-al (4).
4 was
prepared from ent-27 in 72% yield following the procedure
described above for 3. [a]2D4.3 -2.1 (c 0.52, CHCl3).
3-(2-Deoxy-b-D-erythro-pentofuranosyl)-7-(1R,2S-dihydroxy-
heptyl)-3,4-dihydro-9H-imidazo[1,2-a]purin-9-one (13). Nucleo-
side 13 was obtained as a white solid in 62% yield and >99% purity
by analytical HPLC (gradient system II) from (2R,3S,4S)-EHN
(5) following the procedure described above for 11. dH (600 MHz,
DMSO-d6) 0.84 (t, 3H, J = 6.9 Hz, CH3), 1.21–1.32 (m, 6H, H12b,
H13b, 2CH2), 1.45–1.50 (m, 1H, H13a), 1.68–1.73 (m, 1H, H12a),
2.23–2.26 (m, 1H, H2¢¢), 2.58–2.61 (m, 1H, H2¢), 3.49–3.53 (m,
1H, H5¢¢), 3.56–3.60 (m, 1H, H5¢), 3.67 (q, 1H, J = 6.7 Hz, H11),
3.85 (q, 1H, J = 3.8 Hz, H4¢), 4.37–4.39 (m, 2H, H3¢, 11-OH), 4.77
(t, 1H, J = 7.9 Hz, H10), 4.94 (t, 1H, J = 5.4 Hz, 5¢-OH), 5.28 (d,
1H, J = 4.0 Hz, 3¢-OH), 5.35 (d, 1H, J = 8.5 Hz, 10-OH), 6.23 (dd,
1H, J = 6.4, 7.4 Hz, H1¢), 7.23 (s, 1H, H6), 8.12 (s, 1H, H2); dC
(150.9 MHz, DMSO-d6) 154.0, 149.7, 146.7, 137.5, 125.4, 116.1,
115.3, 87.7, 83.0, 72.5, 70.7, 69.8, 61.7, 40.0, 32.8, 31.5, 24.9, 22.1,
13.9. m/z (FAB) 422.2020 (MH+, C19H28N5O6 requires 422.2040)
(2R,3S,4S)-2,3-Epoxy-4-hydroxy-2E-nonan-1-al (5).
5 was
prepared from 28 in 89% yield following the procedure described
above for 3. [a]2D4.3 -40.2 (c 0.61, CHCl3). dH (400 MHz, CD2Cl2);
0.91 (t, 3H, J = 6.8 Hz, CH3), 1.26–1.36 (m, 6H, 3CH2), 1.59–1.65
(m, 2H, CH2), 1.76 (d, 1H, J = 6.7 Hz, OH), 3.28 (dd, 1H, J = 2.0,
4.4 Hz, H3), 3.38 (dd, 1H, J = 2.0, 6.1 Hz, H2), 3. 62 (ddd, 1H,
J = 1.6, 4.7, 6.6 Hz, H4), 9.08 (d, 1H, J = 6.1 Hz, CHO); dC NMR
(75 MHz, CD2Cl2) 198.2, 70.5, 59.8, 57.2, 34.7, 32.0, 25.3, 22.9,
14.1; m/z (GC-EI) 173 (0.2, M++1), 83 (65), 71 (80), 55 (100).
(2S,3R,4R)-2,3-Epoxy-4-hydroxy-2E-nonan-1-al (6).
6 was
prepared from ent-28 in 92% yield following the procedure
described above for 3. [a]2D4 +38.5 (c 0.52, CHCl3).
3-(2-Deoxy-b-D-erythro-pentofuranosyl)-7-(1S,2S-dihydroxy-
heptyl)-3,4-dihydro-9H-imidazo[1,2-a]purin-9-one (11). K2CO3
(8.10 mg, 58.7 mmol) was added to a solution of dGuo·H2O
(7.38 mg, 25.9 mmol) in DMF (500 mL) and the mixture was
stirred for 15 min. A solution of (2S,3R,4S)-EHN (3, 13.40 mg,
77.9 mmol) in DMF (800 mL) was added to the suspension. The
reaction mixture was stirred at room temperature for 12 h; the
formation of the desired product was monitored by HPLC using
gradient systems I (flow rate 1.5 mL min-1). Water was added (4
mL) and reaction mixture was neutralized to pH 7.0 with 1% HCl
(aq). The resulting solution was filtered through a 0.45 mm filter
cartridge prior for purification by semi-preparative HPLC using
gradient system I (flow rate 5.0 mL min-1). The solution from the
reaction was kept frozen during the purification. The fractions
collected from the HPLC were immediately cooled to -78◦ C to
prevent epimerization of the product. Lyophilization afforded the
modified nucleoside 11 (6.55 mg, 60% yield) as a white solid. The
purity of the product was judged to be >99% by analytical HPLC
(gradient system II, flow rate 1.5 mL min-1). dH NMR (600 MHz,
D2O/CD3CN); 0.73 (t, 3H, J = 6.8 Hz, CH3), 1.10–1.30 (m, 5H,
H13b, H14, H15), 1.32–1.41 (m, 3H, H13a, H12), 2.34–2.38 (m,
1H, H2¢¢), 2.65–2.70 (m, 1H, H2¢), 3.63 (dd, 1H, J = 4.1, 12.3 Hz,
H5¢¢), 3.70 (dd, 1H, J = 3.5, 12.3 Hz, H5¢), 3.87–3.91 (m, 1H,
H11), 3.97–3.99 (m, 1H, H4¢), 4.48–4.51 (m, 1H, H3¢), 5.02 (d,
1H, J = 5.5 Hz, H10), 6.22 (t, 1H, J = 6.9 Hz, H1¢), 7.16 (s,
1H, H6), 7.95 (s, 1H, H2); dC (150.9 MHz, D2O/CD3CN) 155.4,
150.3, 147.4, 139.7, 125.6, 117.1, 116.5, 88.3, 85.5, 74.1, 72.3, 70.3,
62.8, 39.8, 33.7, 32.2, 25.8, 23.0, 14.3; m/z (FAB) 422.2020 (MH+,
C19H28N5O6 requires 422.2040).
3-(2-Deoxy-b-D-erythro-pentofuranosyl)-7-(1S,2R-dihydroxy-
heptyl)-3,4-dihydro-9H-imidazo[1,2-a]purin-9-one
(14).
Nucleoside 14 was prepared from (2S,3R,4R)-2,3-epoxy-4-
hydroxynonanal (6) in 71% yield as a white solid and >99% purity
by analytical HPLC (gradient system II) following the procedure
described above for 11. dH (600 MHz, D2O/CD3CN); d 0.79 (t,
3H, J = 6.9 Hz, CH3), 1.16–1.32 (m, 6H, H13b, H12b, 2CH2),
1.40–1.45 (m, 1H, H13a), 1.68–1.75 (m, 1H, H12a), 2.23–2.30
(m, 1H, H2¢), 2.61–2.73 (m, 1H, H2¢), 3.65 (dd, 1H, J = 4.4,
12.4 Hz, H5¢¢), 3.70 (dd, 1H, J = 3.4, 12.4 Hz, H5¢), 3.92–3.89
(m, 1H, H11), 3.99 (q, 1H, J = 3.6, H4¢), 4.37–4.49 (quintet, 1H,
J = 3.0 Hz, H3¢), 4.91 (d, 1H, J = 7.0 Hz, H10), 6.23 (t, 1H, J =
7.0 Hz, H1¢), 7.15 (s, 1H, H6), 7.93 (s, 1H, H2); dC (150.9 MHz,
D2O/CD3CN) 155.6, 150.3, 147.5, 139.8, 125.3, 117.3, 117.0,
88.3, 85.6, 74.0, 72.3, 70.9, 62.9, 39.9, 33.0, 32.3, 25.8, 23.2, 14.4;
m/z (FAB) 422.2018 (MH+, C19H28N5O6 requires 422.2040).
General procedure for synthesis of the tetracyclic
adducts:
(5aR,7S,8S,8aS)-[3-(2-deoxy-b-D-erythro-pentofur-
anosyl)-3,4,5a,7,8,8a-hexahydro-8-hydroxy-7-pentyl-10H-furo[2¢,
3¢:4,5]imidazo[1,2-a]purin-10-one (15) and (5aS,7S,8S,8aR)-[3-
(2-deoxy-b-D-erythro-pentofuranosyl)-3,4,5a,7,8,8a-hexahydro-8-
hydroxy-7-pentyl-10H-furo[2¢,3¢:4,5]imidazo[1,2-a]purin-10-one
(16). Nucleoside 11 (1.9 mg, 4.5 mmol) was dissolved in
phosphate buffer (19 mL, 0.025 M, pH 6.8) and sonicated for
10 min. The reaction mixture was stirred at room temperature for
68 h; during this time, the reaction was monitored by analytical
HPLC (every ~24 h) for the formation of desired products
using gradient system III. The product mixture was filtrated
through a 0.45 mm filter cartridge, rinsed with water (3 ¥ 2
mL), and lyophilized. The resulting solid residue was dissolved
in water (3 mL) and purified by semi-preparative HPLC using
gradient systems IV (flow rate 5.0 mL min-1). The fractions were
3-(2-Deoxy-b-D-erythro-pentofuranosyl)-7-(1R,2R-dihydroxy-
heptyl)-3,4-dihydro-9H-imidazo[1,2-a]purin-9-one (12). Nucleo-
side 12 was obtained from (2R,3S,4R)-EHN (4) as a white solid in
67% yield and >99% purity by analytical HPLC (gradient system
II) following the procedure described above for 11. dH (600 MHz,
D2O/CD3CN) 0.74 (t, 3H, J = 6.9 Hz, CH3), 1.13–1.20 (m, 5H,
1968 | Org. Biomol. Chem., 2011, 9, 1960–1971
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
immediately frozen after collection to prevent the epimerization
of the products and lyophilized to afford nucleosides 15 (1.0 mg,
51% isolated yield; 69% based on recovered starting material) and
16 (0.2 mg, 9% isolated yield; 12% based on recovered starting
material) as white solids. The purity of the obtained adducts were
judged to be >99% by analytical HPLC (gradient systems I or II,
flow rate 1.5 mL min-1). 15: dH (600 MHz, CD3CN/D2O) 0.80
(t, 3H, J = 6.9 Hz, CH3), 1.18–1.31 (m, 6H, 3CH2), 1.59 (q, 2H,
J = 7.3 Hz, CH2 H11), 2.29–2.32 (m, 1H, H2¢), 2.56–2.58 (m, 1H,
H2¢¢), 3.62 (dd, 1H, J = 3.7, 12.4 Hz, H5¢¢), 3.67 (dd, 1H, J = 3.4,
12.4 Hz, H5¢), 3.71 (td, 1H, J = 2.7, 6.9 Hz, H7), 3.96 (q, 1H,
J = 3.2 Hz, H4¢), 4.33 (d, 1H, J = 2.7 Hz, H8), 4.45 (quintet, 1H,
J = 2.8 Hz, H3¢), 4.72 (d, 1H, J = 6.6 Hz, H8a), 5.92 (d, 1H, J =
6.6 Hz, H5a), 6.14 (dd, 1H, J = 6.4, 7.7 Hz, H1¢), 7.82 (s, 1H, H2);
m/z (FAB) 422.2036 (MH+, C19H28N5O6 requires 422.2040).
16: dH (600 MHz, CD3CN/D2O) 0.83 (t, 3H, J = 6.9 Hz, CH3),
1.23–1.36 (m, 6H, 3CH2), 1.54–1.62 (m, 2H, CH2 H11), 2.29–2.33
(m, 1H, H2¢), 2.59–2.63 (m, 1H, H2¢¢), 3.62 (dd, 1H, J = 3.9,
12.4 Hz, H5¢¢), 3.68 (dd, 1H, J = 3.5, 12.4 Hz, H5¢), 3.76 (td, 1H,
J = 2.7, 6.9 Hz, H7), 3.96 (q, 1H, J = 3.3, H4¢), 4.43 (dd, 1H, J =
2.7, 5.8 Hz, H8), 4.45 (quintet, 1H, J = 2.9 Hz, H3¢), 4.97 (dd, 1H,
J = 5.8, 7.7 Hz, H8a), 5.59 (d, 1H, J = 7.7 Hz, H5a), 6.15 (dd, 1H,
J = 6.4, 7.6 Hz, H1¢), 7.83 (s, 1H, H2); m/z (FAB) 422.2052 (MH+,
C19H28N5O6 requires 422.2040).
(5aR,7S,8R,8aS)-[3-(2-deoxy-b-D-erythro-pentofuranosyl)-3,
4,5a,7,8,8a-hexahydro-8-hydroxy-7-pentyl-10H-furo[2¢,3¢:4,5]-
imidazo[1,2-a]purin-10-one (19) and (5aS,7S,8R,8aR)-[3-(2-deoxy-
b-D-erythro-pentofuranosyl)-3,4,5a,7,8,8a-hexahydro-8-hydroxy-
7-pentyl-10H-furo[2¢,3¢:4,5]imidazo[1,2-a]purin-10-one (20). Fol-
lowing the general procedure described above, 13 was (2.9 mg, 6.8
mmol) was incubated in phosphate buffer (29 mL, 0.025 M, pH 6.8)
for 114 h. Purification by semi-preparative HPLC using gradient
system V (flow rate 5.0 mL min-1 afforded 19 (0.7 mg, 24% isolated
yield; 30% based on recovered starting material) and 20 (1.1 mg,
38% isolated yield; 49% based on recovered starting material) as
white solids. The purity of the products was judged to be >99%
by analytical HPLC (gradient system I or II).
19: dH (600 MHz, CD3CN/D2O) 0.82 (t, 3H, J = 6.1 Hz, CH3),
1.20–1.25 (m, 4H, 2CH2), 1.32–1.40 (m, 2H, H12a, H12b), 1.45–
1.52 (m, 1H, H11b), 1.60–1.66 (m, 1H, H11a), 2.30–2.34 (m, 1H,
H2¢¢), 2.56–2.60 (m, 1H, H2¢), 3.58 (dt, 1H, J = 4.3, 8.0 Hz, H7),
3.61 (dd, 1H, J = 4.0, 12.4 Hz, H5¢¢), 3.71 (dd, 1H, J = 3.5, 12.3 Hz,
H5¢), 3.97 (m, 1H, H4¢), 4.07 (dd, 1H, J = 7.1, 7.9 Hz, H8), 4.45 (q,
1H, J = 2.9 Hz, H3¢), 5.0 (t, 1H, J = 7.0 Hz, H8a), 5.89 (d, 1H, J =
6.9 Hz, H5a), 6.14 (dd, 1H, J = 6.4, 7.5 Hz, H1¢), 7.83 (s, 1H, H2);
m/z (ESI-TOF) 422.2032 (MH+, C19H28N5O6 requires 422.2040).
20: dH (600 MHz, CD3CN/D2O) 0.75 (t, 3H, J = 7.0 Hz, CH3),
1.12–1.16 (m, 4H, 2CH2), 1.20–1.23 (m, 2H, H11b, H12b), 1.30–
1.33 (m, 2H, H11a, H12a), 2.28–2.33 (m, 1H, H2¢), 2.56–2.60 (m,
1H, H2¢¢), 3.62 (dd, 1H, J = 4.0, 12.3 Hz, H5¢¢), 3.67 (dd, 1H, J =
3.6, 12.4 Hz, H5¢), 3.90 (ddd, 1H, J = 2.6, 5.4, 8.1 Hz, H7), 3.95
(q, 1H, J = 3.4 Hz, H4¢), 4.33 (dd, 1H, J = 1.2, 2.6 Hz, H8), 4.45
(q, 1H, J = 2.9 Hz, H3¢), 4.73 (dd, 1H, J = 1.2, 6.7 Hz, H8a), 5.86
(d, 1H, J = 6.7 Hz, H5a), 6.14 (dd, 1H, J = 6.6, 7.5 Hz, H1¢), 7.83
(s, 1H, H2); m/z (ESI-TOF) 422.2037 (MH+, C19H28N5O6 requires
422.2040), 444.1859 (M+Na, C19H27N5O6Na requires 444.1859).
(5aS,7R,8R,8aR)-[3-(2-deoxy-b-D-erythro-pentofuranosyl)-
3,4,5a,7,8,8a-hexahydro-8-hydroxy-7-pentyl-10H-furo[2¢,3¢:4,5]-
imidazo[1,2-a]purin-10-one (17) and (5aR,7R,8R,8aS)-[3-(2-
deoxy-b-D-erythro-pentofuranosyl)-3,4,5a,7,8,8a-hexahydro-8-
hydroxy-7-pentyl-10H-furo[2¢,3¢:4,5]imidazo[1,2-a]purin-10-one
(18)
Following the general procedure as described above, 12 (6.9 mg,
16.8 mmol) was incubated in phosphate buffer (69 mL, 0.025 M,
pH 6.8) for 96 h to afford 17 (4.1 mg, 58% isolated yield; 68% based
on recovered starting material) and 18 (0.65 mg, 9% isolated yield;
10% based on recovered starting material) were obtained as white
solids. The purity of the products was judged to be >99.9% by
analytical HPLC (gradient system I or II).
17: UV lmax (5 : 1 H2O/CH3CN) nm 250 (e 12,908); dH
(600 MHz, CD3CN/D2O) 0.78 (t, 3H, J = 6.8 Hz, CH3), 1.18–
1.26 (m, 6H, 3CH2), 1.58 (q, 2H, J = 7.3 Hz, CH2 H11), 2.29–2.34
(m, 1H, H2¢), 2.56–2.60 (m, 1H, H2¢¢), 3.61 (dd, 1H, J = 4.1 Hz,
J = 12.4 Hz, H5¢¢), 3.67 (dd, 1H, J = 3.6 Hz, J = 12.4 Hz, H5¢),
3.71 (td, 1H, J = 2.6 Hz, J = 6.9 Hz, H7), 3.95 (q, 1H, J = 3.5 Hz,
H4¢), 4.33 (d, 1H, J = 2.6 Hz, H8), 4.45 (quintet, 1H, J = 2.9 Hz,
H3¢), 4.72 (d, 1H, J = 6.6 Hz, H8a), 5.91 (d, 1H, J = 6.6 Hz, H5a),
6.16 (dd, 1H, J = 6.3 Hz, J = 7.7 Hz, H1¢), 7.83 (s, 1H, H2); m/z
(FAB) 422.2063 (MH+, C19H28N5O6 requires 422.2040).
18: dH (600 MHz, CD3CN/D2O) 0.82 (t, 3H, J = 6.9 Hz, CH3),
1.24–1.35 (m, 6H, 3CH2), 1.55–1.60 (m, 2H, J = 6.9 Hz, CH2,
H11), 2.30–2.35 (m, 1H, H2¢), 2.57–2.62 (m, 1H, H2¢¢), 3.62 (dd,
1H, J = 4.0, 12.4 Hz, H5¢¢), 3.68 (dd, 1H, J = 3.5, 12.4 Hz, H5¢),
3.76 (td, 1H, J = 2.6, 6.9 Hz, H7), 3.96 (q, 1H, J = 3.1 Hz, H4¢),
4.43 (dd, 1H, J = 2.5 Hz, J = 5.8 Hz, H8), 4.45 (quintet, 1H, J =
2.9 Hz, H3¢), 4.98 (dd, 1H, J = 5.8, 7.6 Hz, H8a), 5.59 (d, 1H,
J = 7.7 Hz, H5a), 6.15 (dd, 1H, J = 6.5, 7.4 Hz, H1¢), 7.84 (s,
1H, H2); m/z (ESI-TOF) 422.2056 (MH+, C19H28N5O6 requires
422.2040).
(5aS,7R,8S,8aR)-[3-(2-deoxy-b-D-erythro-pentofuranosyl)-3,
4,5a,7,8,8a-hexahydro-8-hydroxy-7-pentyl-10H-furo[2¢,3¢:4,5]-
imidazo[1,2-a]purin-10-one (21) and (5aR,7R,8S,8aS)-[3-(2-deoxy-
b-D-erythro-pentofuranosyl)-3,4,5a,7,8,8a-hexahydro-8-hydroxy-
7-pentyl-10H-furo[2¢,3¢:4,5]imidazo[1,2-a]purin-10-one (22). Fol-
lowing the general procedure described above, 14 (2.5 mg, 5.9
mmol) was incubated in phosphate buffer (24 mL, 0.025 M, pH 6.8)
for 72 h. Purification by semi-preparative HPLC using gradient
system V (flow rate 5.0 mL min-1) afforded 21 (0.5 mg, 19% isolated
yield; 23% based on recovered starting material) and 22 (0.75 mg,
29% isolated yield; 36% based on recovered starting material) as
white solids. The purity of the products was judged to be >99.5%
by analytical HPLC (gradient system I or II).
21: dH (600 MHz, CD3CN/D2O) 0.85 (t, 3H, J = 6.2 Hz, CH3),
1.27–1.29 (m, 4H, 2CH2), 1.37–1.43 (m, 2H, H12a, H12b), 1.49–
1.55 (m, 1H, H11b), 1.63–1.69 (m, 1H, H11a), 2.33–2.37 (m, 1H,
H2¢¢), 2.62–2.67 (m, 1H, H2¢), 3.59 (dt, 1H, J = 4.3, 8.0, H7), 3.64
(dd, 1H, J = 4.0, 12.4 Hz, H5¢¢), 3.72 (dd, 1H, J = 3.5, 12.4 Hz,
H5¢), 4.05 (m, 1H, H4¢), 4.10 (dd, 1H, J = 7.1, 8.0 Hz, H8), 4.49
(quintet, 1H, J = 2.9 Hz, H3¢), 5.04 (t, 1H, J = 7.0 Hz, H8a), 5.92
(d, 1H, J = 6.9 Hz, H5a), 6.17 (dd, 1H, J = 6.4, 7.6 Hz, H1¢), 7.88
(s, 1H, H2); m/z (ESI-TOF) 444.1855 (M+Na, C19H27N5O6Na
requires 444.1859.
22: dH (600 MHz, CD3CN/D2O) 0.78 (t, 3H, J = 7.1 Hz, CH3),
1.16–1.19 (m, 4H, 2CH2), 1.23–1.26 (m, 2H, H11b, H12b), 1.33–
1.36 (m, 2H, H11a, H12a), 2.32–2.36 (m, 1H, H2¢), 2.58–2.63 (m,
This journal is
The Royal Society of Chemistry 2011
Org. Biomol. Chem., 2011, 9, 1960–1971 | 1969
©

View Article Online
1H, H2¢¢), 3.63 (dd, 1H, J = 3.9, 12.4 Hz, H5¢¢), 3.70 (dd, 1H, J =
3.5, 12.4 Hz, H5¢), 3.92 (ddd, 1H, J = 2.6, 5.3, 7.9 Hz, H7), 3.98–
3.99 (m, 1H, H4¢), 4.32 (dd, 1H, J = 1.2, 2.6 Hz, H8), 4.48 (q, 1H,
J = 2.9 Hz, H3¢), 4.72 (dd, 1H, J = 1.2, 6.7 Hz, H8a), 5.89 (d, 1H,
J = 6.7 Hz, H5a), 6.17 (dd, 1H, J = 6.4, 7.6 Hz, H1¢), 7.87 (s, 1H,
H2); m/z (ESI-TOF) 444.1858 (M+Na, C19H27N5O6Na requires
444.1859).
cooled NMR probe. Chemical shifts were referenced internally to
CD3CN (1.93 ppm) or DMSO-d6 (2.49 ppm), which also served
2
as the H lock solvents. Typical experimental conditions for 1D
1H NMR spectra included 32 K data points, 13 ppm sweep width,
a recycle delay of 1.5 s and 32–512 scans depending on sample
concentration. Experimental conditions for 2D 1H–1H COSY
´
analysis included 2048 I 512 data matrix, 13 ppm sweep width,
recycle delay of 1.5 s and 4 scans per increment. The data was
processed using squared sinebell window function, symmetrized,
and displayed in magnitude mode.
HPLC. The analysis of reaction mixtures and the purification
of nucleosides were conducted on a gradient HPLC system with a
diode array UV detector monitoring at 260 nm using C18 reversed-
phase columns (250 ¥ 4.6 mm i.d., flow rate of 1.5 mL min-1
for analysis and 250 ¥ 10 mm i.d, flow rate 5 mL min-1 for
purification). The mobile phase consisted of H2O and CH3CN
using the following gradients: Gradient I: initially 99% H2O; a
15 min linear gradient to 90% H2O; a 5 min linear gradient to 80%
H2O; isocratic at 80% H2O for 5 min, 10 min linear gradient to
20% H2O, isocratic at 20% H2O for 5 min, followed by a 5 min
linear gradient to the initial conditions. Gradient II: initially 90%
H2O; a 25 min linear gradient to 20% H2O, followed by a 5 min
linear gradient to the initial conditions. Gradient III: initially 90%
H2O; a 5 min linear gradient to 75% H2O; isocratic at 75% H2O
for 20 min, 4 min linear gradient to 20% H2O, followed by a 4 min
linear gradient to the initial conditions. Gradient IV: initially 90%
H2O; a 4 min linear gradient to 75% H2O; isocratic at 75% H2O
for 20 min, 4 min linear gradient to 20% H2O; isocratic at 80%
H2O for 2 min, followed by a 4 min linear gradient to the initial
conditions. Gradient V: initially 90% H2O; a 3 min linear gradient
to 78% H2O; isocratic at 78% H2O for 25 min, 4 min linear gradient
to 20% H2O; isocratic at 80% H2O for 3 min, followed by a 4 min
linear gradient to the initial conditions.
Double Pulse Field Gradient Spin Echo (DPFGSE)–NOE
and homonuclear decoupling NMR experiments. Selective 1D
NOE spectra were recorded using the DPFGSE technique.43–45
Experimental parameters for this experiment were similar to
1
those for the standard 1D H experiment with the addition of
600 ms mixing time. A standard Gaussian shaped pulse was
used for the excitation pulse; the operating software automatically
calculated the length and power of the pulse based on the peak
integration area of the signal that was irradiated. All samples
were dissolved in D2O/CD3CN (1 : 1) and degassed by carefully
bubbling high purity He gas through the solution in the NMR
tube for ~10 min. A fine capillary was used to slowly introduce
the He gas to avoid blowing the sample out of the tube. The
1
1D H homonuclear decoupling experiments were run using the
standard pulse sequence included in the manufacturer’s software;
the software automatically calculated the irradiation frequencies.
Acknowledgements
The National Institutes of Health supported this work through re-
search grants P01 ES11331 and R01 ES05355 and center grant P30
ES00267. The National Center for Research Resources provided
funding for the 600 MHz NMR instrument (S10 RR019022). Dr
Ivan Kozekov is acknowledged for many helpful suggestions.
Separation of adducts 10–22. A solution containing adducts
11–22 was prepared from standard solutions of the pure com-
pounds (1.0 mg mL-1 in 10 : 1 H2O/CH3CN). The mixture was
analyzed by reversed-phase HPLC (250 mm ¥ 4.6 mm, flow rate
1.0 mL min-1). The mobile phase consisted of H2O and CH3CN
using the following gradients: initially 90% H2O; a 3 min linear
gradient to 78% H2O; isocratic at 78% H2O for 25 min, 4 min
linear gradient to 20% H2O, followed by a 4 min linear gradient
to 90% H2O and remained for 1 min at the initial conditions.
Notes and references
1 N. A. Porter, S. E. Caldwell and K. A. Mills, Lipids, 1995, 30, 277–290.
2 H. Esterbauer, R. J. Schaur and H. Zollner, Free Radical Biol. Med.,
1991, 11, 81–128.
3 S. H. Lee and I. A. Blair, Trends Cardiovasc. Med., 2001, 9, 148–155.
4 K. Uchida, Free Radical Biol. Med., 2000, 28, 1685–1696.
5 P. C. Burcham, Mutagenesis, 1998, 13, 287–305.
6 G. Spiteller, Chem. Phys. Lipids, 1998, 95, 105–162.
7 K. Uchida, Prog. Lipid Res., 2003, 42, 318–343.
8 C. Ji, V. Amarnath, J. A. Pietenpol and L. J. Marnett, Chem. Res.
Toxicol., 2001, 14, 1090–1096.
9 J. D. West, C. Ji, S. T. Duncan, V. Amarnath, C. Schneider, C. J. Rizzo,
A. R. Brash and L. J. Marnett, Chem. Res. Toxicol., 2004, 17, 453–462.
10 C. K. Winter, H. J. Segall and W. F. Haddon, Cancer Res., 1986, 46,
5682–5686.
Electronic Circular Dichroism (ECD) spectroscopy. All ECD
spectra were recorded at 25 ◦C in a 1.0 cm circular quartz cell
and are an average of three individual scans. The instrument
was calibrated with ammonium d-camphor-10-sulfonate. The
sample concentrations were 0.025–0.05 mg mL-1 in 5 : 1 (v/v)
H2O/CH3CN. The spectra were recorded at a scan speed of
100 nm min-1 and a time constant of 1s. The parameters
bandwidth of 1 nm, resolution of 1 nm, and sensitivity of 100
mdeg were fixed before recording the spectra. The corresponding
solvent spectrum was subtracted from the ECD spectrum of
nucleoside solution. The resulting spectra were further processed
for smoothening when needed, using the manufacturer’s software.
The ECD intensities are presented in units of molar ellipticity (H)
vs. wavelength.
11 P. Yi, D. Zhan, V. M. Samokyszyn, D. R. Doerge and P. P. Fu, Chem.
Res. Toxicol., 1997, 10, 1259–1265.
12 F. L. Chung, R. G. Nath, J. Ocando, A. Nishikawa and L. Zhang,
Cancer Res., 2000, 60, 1507–1511.
13 H. Wang, I. D. Kozekov, T. M. Harris and C. J. Rizzo, J. Am. Chem.
Soc., 2003, 125, 5687–5700.
14 P. H. Fernandes, H. Wang, C. J. Rizzo and R. S. Lloyd, Environ. Mol.
Mutagen., 2003, 42, 68–74.
15 R. S. Sodum and F. L. Chung, Chem. Res. Toxicol., 1989, 2, 23–28.
16 R. S. Sodum and F. L. Chung, Cancer Res., 1991, 51, 137–143.
17 T. Douki, F. Odin, S. Caillat, A. Favier and J. Cadet, Free Radical Biol.
Med., 2004, 37, 62–70.
NMR experiments for the nucleosides. NMR experiments were
recorded using a 14.0 T Bruker magnet equipped with Bruker AV-
III console operated at 600.13 MHz. All spectra were acquired
in 3 mm NMR tubes using a Bruker 5 mm TCI cryogenically
18 R. S. Sodum and F. L. Chung, Cancer Res., 1988, 48, 320–323.
1970 | Org. Biomol. Chem., 2011, 9, 1960–1971
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
19 F. L. Chung, H.-J. C. Chen, J. B. Guttenplan, A. Nishikawa and G. C.
Hard, Carcinogenesis, 1993, 14, 2073–2077.
20 H.-J. C. Chen and F.-L. Chung, Chem. Res. Toxicol., 1996, 9, 306–312.
21 S. H. Lee, T. Oe and I. A. Blair, Science, 2001, 292, 2083–2086.
22 B. L. Golding, P. K. Slaich, G. Kennedy, C. Bleasdale and W. P. Watson,
Chem. Res. Toxicol., 1996, 9, 147–157.
23 V. Nair and R. J. Offerman, J. Org. Chem., 1985, 50, 5627–5631.
24 B. T. Golding, P. K. Slaich and W. P. Watson, IARC Sci Publ, 1986, 70,
227–231.
33 S.-J. Chang, D. McNally, S. Shary-Tehrany, M. J. Hickey and R. H.
Boyd, J. Am. Chem. Soc., 1970, 92, 3109–3118.
34 J. W. Barrett and R. P. Linstead, J. Chem. Soc., 1936, 611–616.
35 C. A. G. Haasnoot, F. A. A. M. DeLeeuw and C. Altona, Tetrahedron,
1980, 36, 2783–2792.
36 L. V. Nechev, C. M. Harris and T. M. Harris, Chem. Res. Toxicol., 2000,
13, 421–429.
37 A. K. Brock, I. D. Kozekov, C. J. Rizzo and T. M. Harris, Chem. Res.
Toxicol., 2004, 17, 1047–1056.
25 H.-J. C. Chen, L. Zhang, J. Cox, J. A. Cunningham and F.-L. Chung,
Chem. Res. Toxicol., 1998, 11, 1474–1480.
38 C. Mesaros, B. G. Gugiu, R. Zhou, S. H. Lee, J. Choi, J. Laird, I. A.
Blair and R. G. Salomon, Chem. Res. Toxicol., 2010, 23, 516–527.
39 S. H. Lee, J. A. Arora, T. Oe and I. A. Blair, Chem. Res. Toxicol., 2005,
18, 780–786.
40 S. H. Lee, M. V. Silva Elipe, J. S. Arora and I. A. Blair, Chem. Res.
Toxicol., 2005, 18, 566–578.
26 L. Yu and Z. Wang, J. Chem. Soc., Chem. Commun., 1993, 232–234.
27 Y. Gao, R. M. Hanson, J. M. Klunder, S. Y. Ko, H. Masamune and
K. B. Sharpless, J. Am. Chem. Soc., 1987, 109, 5765–5780.
28 P. Allevi, M. Anastasia, F. Cajone, P. Ciuffreda and A. M. Sanvito,
J. Org. Chem., 1993, 58, 5000–5002.
41 S. H. Lee, T. Oe and I. A. Blair, Chem. Res. Toxicol., 2002, 15, 300–
304.
29 L. de Montarby, P. Mosset and R. Gree, Tetrahedron Lett., 1988, 29,
3937–3940.
42 A. P. M. Loureiro, P. Di Mascio, O. F. Gomes and M. H. G. Medeiros,
Chem. Res. Toxicol., 2000, 13, 601–609.
30 A. De Mico, R. Margarita, L. Parlanti, A. Vescovi and G. Piancatelli,
J. Org. Chem., 1997, 62, 6974–6977.
31 K. V. Petrova, R. S. Jalluri, I. D. Kozekov and C. J. Rizzo, Chem. Res.
Toxicol., 2007, 20, 1685–1692.
32 A. P. M. Loureiro, I. P. de Arruda Campos, O. F. Gomes, E. P. M.
Possari, P. Di Mascio and M. H. G. Medeiros, Chem. Res. Toxicol.,
2005, 18, 290–299.
43 T. L. Hwang and A. J. Shaka, J. Magn. Reson., Ser. A, 1995, 112,
275–279.
44 K. Stott, J. Keeler, Q. N. Van and A. J. Shaka, J. Magn. Reson., 1997,
125, 302–324.
45 K. Stott, J. Stonehouse, J. Keeler, T.-L. Hwang and A. J. Shaka, J. Am.
Chem. Soc., 1995, 117, 4199–4200.
This journal is
The Royal Society of Chemistry 2011
Org. Biomol. Chem., 2011, 9, 1960–1971 | 1971
©