Site-specific incorporation of the 1-hexanol-1,N6-etheno-2′-deoxyadenosine adduct into oligodeoxyribonucleotides

Article abstract of DOI:10.1016/S0968-0896(03)00125-1

Modified oligonucleotides that contain the hydrophobic 1-hexanol-1,N⁶-etheno-2′-deoxyadenosine adduct have been synthesized using a mild solid phase phosphoramidite chemistry. The presence and the integrity of the modified nucleoside in the syn

Full text of DOI:10.1016/S0968-0896(03)00125-1

Bioorganic & Medicinal Chemistry 11 (2003) 2445–2452
Site-Specific Incorporation of the 1-Hexanol-1,N6-etheno-20-
deoxyadenosine Adduct into Oligodeoxyribonucleotides
Valdemir M. Carvalho,a Didier Gasparutto,b Paolo Di Mascio,a
Marisa H. G. Medeirosa and Jean Cadetb,*
a
´
Departamento de Bioquımica, Instituto de Quımica, Universidade de Sa˜o Paulo, CP 26 077, CEP 05513-970, Sa˜o Paulo, Brazil
´
b
´ ´
Laboratoire Lesions des Acides Nucleiques, DRFMC/SCIB-FRE 2600, CEA/Grenoble, 38054 Grenoble Cedex 9, France
Received 18 October 2002; accepted 18 February 2003
Abstract—Modified oligonucleotides that contain the hydrophobic 1-hexanol-1,N6-etheno-20-deoxyadenosine adduct have been
synthesized using a mild solid phase phosphoramidite chemistry. The presence and the integrity of the modified nucleoside in the
synthetic oligomers were confirmed by electrospray ionization and MALDI mass spectrometry measurements together with analysis
of the complete enzymatic hydrolysate by high performance liquid chromatography coupled to UV and fluorescent detection
techniques.
# 2003 Elsevier Science Ltd. All rights reserved.
Introduction
supported by several studies.6,7 Thus, trans-4-hydroxy-
2-nonenal (HNE), one of the most studied breakdown
LP products, is able to react with peroxides to give rise
to an epoxide derivative. The latter compound may
rearrange prior to generate 1,N6-edAdo, 1,N2-etheno-20-
deoxyguanosine (1,N2-edGuo) and their respective
dihydroxyheptyl derivatives.8,9 We have also demon-
strated that trans-2-octenal and trans,trans-2,4-deca-
dienal (DDE) are efficient DNA alkylating agents. Thus,
DDE is indeed very versatile being able to produce,
after epoxidation, the non substituted DNA adducts
formed by HNE, together with three other products,
namely decan-1,2,3-triol, 2,3-epoxy-1-decanol and
1-hexanol-1,N6-edAdo.10,11 It has been recently shown
that trans-4-oxo-2-noneal, another lipid peroxidation
product, also induces the formation of substituted
ethenoadducts.12 Each a,b-unsaturated aldehyde arising
from the breakdown of lipid peroxides appears to form
one or more different substituted ethenoadducts. Over-
all, the adducts share some common features: they are
bulky modifications and the alkyl side chain attached to
the etheno ring confers high hydrophobicity. They differ
mostly in terms of side-chain length and the presence of
hydroxyl groups. Among the already characterized
substituted ethenoadducts, 1-[3-(2-deoxy-b-d-erythro-
pentofuranosyl)-3H-imidazo-[2,1-i]-purin-7-yl]-1-hexanol
(1-hexanol-edAdo) 1 generated from the reaction of
20-deoxyadenosine with either trans-2-octenal or DDE11
has the simplest structure.
Ethenoadducts (e) to DNA appear to be key lesions to
understand the association between many carcinogenic
compounds and their ominous effects. They were initi-
ally reported to be the main adducts arising from the
reaction of chloroacetaldehyde with amino substituted
nucleobases.1 Later, it was shown that metabolites from
occupational carcinogens such as vinyl halides, vinyl
carbamate and urethane also react with DNA to form
the latter exocyclic adducts. These findings motivated a
number of studies aimed at evaluating the mutagenic
properties of these DNA lesions (see ref 2 for a review).
For this purpose, 1,N6-etheno-200-deoxyadenosine adduct
(edAdo) and 3,N4-etheno-2 -deoxycytidine adduct
(edCyd) were transfected in host cells as site-specifically
modified plasmids within shuttle vectors. Then, after
replication in mammalian cells, high frequent mutagenic
events were found to occur, with yields of 81 and 70%,
respectively, at defined positions.3,4
The detection of background levels of edAdo and
edCyd in tissues from unexposed rats and humans sug-
gested an endogenous pathway for the formation of
these adducts.5 Suggestion that lipid peroxidation (LP)
is the endogenous source of ethenoadduct formation is
*Corresponding author. Tel.: +33-4-3878-4987; fax: +33-4-3878-
5090; e-mail: cadet@drfmc.ceng.cea.fr
0968-0896/03/$ - see front matter # 2003 Elsevier Science Ltd. All rights reserved.
doi:10.1016/S0968-0896(03)00125-1

2446
V. M. Carvalho et al. / Bioorg. Med. Chem. 11 (2003) 2445–2452
In order to investigate the biological features of this
kind of DNA damage, site-specific incorporation of the
targeted modification into oligonucleotides is necessary.
Therefore, we reported here an efficient synthesis of the
1-hexanol-edAdo adduct 1 together with a protective
strategy to prepare its phosphoramidite building block
(Scheme 1). The latter phosphoramidite derivative was
successfully site-specifically incorporated into defined
oligodeoxynucleotides by mild chemical DNA synthesis
on solid support.
1-hexanol-edA 1 appears to be a good candidate to
delineate the biological properties of a substituted
ethenoadduct in term of replication, mutagenesis, and
repair. Thus, we have focused our attention on the
development of a strategy aimed at incorporating 1 into
oligonucleotide sequences.
The first requirement to reach this purpose is to syn-
thesize large amounts of the adduct 1 to allow a wide set
of stability assays and to define protection steps. In
order to improve the formation yields of 1, it was
necessary to use a more efficient epoxidation method.
Thus, trans-2-octenal epoxidation derivative was pre-
pared by a method previously described by Yang13 with
some modifications. The approach is well suited for
unsaturated carbonyl compounds through the in situ
generation of the strong epoxidizing trifluoro-
methyldioxirane agent. Reaction of an excess of the
unpurified trans-2-octenal oxidation product with dAdo
allows the formation of 1-hexanol-edAdo 1 in a 60%
yield. Then, 1 was purified by preparative HPLC and
comparison of its spectroscopic properties with those of
an authentic sample11 proved the identity of the product.
Results and Discussion
Synthesis of 1-hexanol-"dAdo 1
Recently, we have reported that 20-deoxyadenosine
(dAdo) was able to react with trans-2-octenal in the
presence of peroxides to produce a main fluorescent pro-
duct that was characterized as 1-hexanol-edAdo 1.11 In the
latter study, trans-2-octenal was shown to be an efficient
alkylating agent yielding ethenoadduct levels comparable
to HNE. Taking into account its high formation yield,
Scheme 1. Synthesis pathway used for the preparation of the phosphoramidite building block of 1-hexanol-edAdo 1: (a) trifluoroacetone, Oxone1
,
CH3CN, aqueous EDTA, 4 h; (b) sodium phosphate buffer, 50 ꢀC, 120 h, 58%; (c) TIPDSCl2 (1.1 equiv), imidazol (4.4 equiv), triethylamine
(2 equiv), DMF, 2 h, 70%; (d) isobutyric anhydride (3.6 equiv), DMAP (6.2 equiv), pyridine, 30 min, 83%; (e) CsF (10 equiv), pyridine, 24 h, 85%;
(f) DMTrCl (1.1 equiv), pyridine, 1 h, 85%; (g) Cl–P(N–iPr2)O(CH2)2CN (1.3 equiv), diisopropylethylamine (2.4 equiv), CH2Cl2, argon, 1 h, 62%.

V. M. Carvalho et al. / Bioorg. Med. Chem. 11 (2003) 2445–2452
2447
Stability studies of 1-hexanol-"dAdo 1
monitoring and UV detection at l278nm confirmed the
homogeneity of 6. The 31P NMR features of purified 6
showed the presence of a 1:1 mixture of diastereo-
isomers (d 149.3 and 149.2 ppm, respectively). The latter
analysis showed also the presence of a very small
amount (less than 5%) of the phosphonate derivatives
(d 8.4 ppm), indicating that the phosphoramidite is
stable during the HPLC purification process.
In order to estimate the stability of 1-hexanol-edAdo 1
during the different oligonucleotide solid phase synth-
esis steps, the modified nucleoside was treated with acid
(80% acetic acid), oxidizing (0.1 M iodine solution in
THF/water) and alkali solutions (32% aqueous ammo-
nia solution and 0.05 M potassium carbonate in metha-
nol). Then, the degradation of 1 was quantified by
reversed phase HPLC using UV and fluorescence
detections. Thus, 1-hexanol-edAdo 1 was found to be
quite stable under acid and oxidizing conditions. How-
ever, as numerous DNA exocyclic adducts, 1 showed a
high unstability in ammonia solutions (complete degra-
dation after 2 h at room temperature). An alternative
deprotection reagent for alkali-labile DNA component-
containing oligonucleotides is methanolic K2CO3.14 The
latter treatment was already successfully applied as a
mild basic deprotection agent for the incorporation of
malonaldialdehyde-20-deoxyguanosine15,16 and p-benzo-
quinone-20-deoxyguanosine adducts17 into DNA frag-
ments. 1-Hexanol-edAdo 1 appeared stable under the
K2CO3 treatment (less than 10% degradation after 24 h
at room temperature), allowing the use of this depro-
tection strategy in combination with the phenoxyacetyl
protecting groups18 for the amino functions of the
adduct 1 and unmodified nucleosides.
Solid-phase
oligonucleotides containing 1-hexanol-"dA 1
synthesis
and
characterization
of
The phosphoramidite synthon was incorporated into
three oligonucleotide sequences (Table 1) following the
solid-phase phosphoramidite method. Commercially
available unmodified 20-deoxyribonucleoside phosphor-
amidite derivatives protected with the phenoxyacetyl
(dAdo and dGuo) and acetyl (dCyd) groups were cho-
sen with regard to their high alkali lability.15,18 By
increasing the coupling time of phosphoramidite syn-
thon 6 from 30 s to 10 min, it was possible to obtain a
coupling yield up to 90% for the modified monomer, as
monitored by the release of the DMTr cation during the
detritylation step. The cleavage and complete deprotec-
tion of the oligonucleotides were accomplished upon
incubation with 50 mM K2CO3 in dry methanol14 for
12 h. The crude 50-DMTr-oligomers were purified and
deprotected on-line with 1% aqueous trifluoroacetic
acid solution by reversed phase HPLC.20 Finally, the
synthetic modified DNA oligomers were further purified
by preparative denaturating polyacrylamide gel electro-
phoresis (PAGE).
Preparation of the protected phosphoramidite building
block of 1-hexanol-"dAdo 1
Besides its unstability, another potential difficulty for
1-hexanol-edAdo 1 incorporation into oligonucleotides
is the presence of an additional hydroxyl group on the
adduct side-chain (Scheme 1). This could lead to unde-
sirable elongation during the oligonucleotide synthesis
cycle. Therefore, it was necessary to protect selectively
this reactive function. The secondary alcohol groups
within the 2-deoxyribose moiety (30-OH) and the etheno
ring (13-OH) showed very similar reactivities. A number
of protecting groups such as acetyl, tert-butyldi-
methylsilyl and levulinyl were unsuccessfully tested
while a mixture of the products substituted at both
positions was obtained (data not shown). This difficulty
was overcome by the simultaneous protection of both
2-deoxyribose hydroxyl groups using the bifunctional
silyl protecting group, namely 1,1,3,3-tetraiso-
propyldisiloxane-1,3-diyl (TIPDS).19 By adding TIPDS
chloride, carefully and in small portions, it was possible
to obtain the desired nucleoside 2, with a protection at
the 30,50-position as the major product (70% yield),
along with a very small amount of the completely silyl-
protected derivative. After achieving the 2-deoxyribose
protection, it was possible to block the 13-OH function
with the isobutyryl group to give rise to derivative 3 in
83% yield. Quantitative removal of the 30,50-O-dis-
iloxane group with cesium fluoride afforded compound
4. Then, the primary 50-OH of 4 was efficiently pro-
tected with a dimethoxytrityl group, yielding product 5.
Finally, the latter nucleoside was phosphitylated using
the standard methodology to afford the expected phos-
phoramidite building block 6, after preparative RP-
HPLC purification. HPLC analysis with fluorescence
The structure of the 5-mer, 14-mer and 22-mer oligo-
nucleotides was assessed by electrospray ionization
(ESI) and MALDI-TOF mass spectrometry measure-
ments (Fig. 1; Table 1). The molecular masses
obtained by both techniques were in agreement with
the calculated molecular masses and therefore con-
firmed the presence and the integrity of the base adduct
within the DNA fragments. The presence of the adduct
1-hexanol-edAdo 1 in the oligonucleotide sequences
received further support from the reversed phase
HPLC analysis coupled to UV and fluorescent detec-
tion of the enzymatic digestion mixture of the modified
DNA fragments as shown in Figure 2. Thus, it
clearly appears that, besides the four normal 20-deoxy-
ribonucleosides, 1-hexanol-edAdo 1 that eluted at
37 min was present. The identity and integrity of 1 were
confirmed by co-injection with a standard together with
ESI-MS measurement of the content of the collected
peak.
Table 1. Sequences and relative molecular masses (Da) of the modified
ODNs synthesized (ESI MS and MALDI-TOF MS measurements were
performed in the negative mode)
Sequences
Mass
calcd
ESI- MALDI-
MS MS
50-d(T[1]GCA)-30
1611.2 1611.3 1611.1
4378.0 4376.0 4377.4
50-d(ATCGTG[1]CTGATCT)-30
50-d(CACTTCGG[1]TCGTGACTGATCT)-30 6825.6 6825.1 6825.0

2448
V. M. Carvalho et al. / Bioorg. Med. Chem. 11 (2003) 2445–2452
Figure 1. Electrospray ionization mass spectrum, in the negative mode, of the modified 14-mer oligonucleotide [seq: 50-d(ATCGTG[1]CTGATCT)-30;
Mr=4378.0 Da).
0
0
Figure 2. HPLC profile of the enzymatic digestion mixture of ODN [5 -d(T[1]GCA)-3 ] with nuclease P1, calf spleen phosphodiesterase, bovine
intestinal mucosa phosphodiesterase and calf intestine alkaline phosphatase. The digestion mixture was analyzed by RP-HPLC (see Experimental
for conditions). The detection was achieved by UV measurement at 260 nm (B) together with specific fluorescence measurement of 1-hexanol-edAdo
(lex=310 nm/lem=420 nm) (A). The inset shows the electrospray ionization mass spectrum, in the positive mode, of the collected 1-hexanol-edAdo
residue 1, eluted at 37 min.
Conclusion
solid-phase assembling, the synthetic oligonucleotides
were isolated in good yields and characterized by mass
spectrometry measurements and complete enzymatic
digestion, showing the integrity of the modified nucleo-
side 1 within the synthetic DNA oligomers. This meth-
odology, that may be applied to the incorporation of
numerous adducts including oxidized and alkylated
DNA modifications,21,22 should allow the assessment of
the biological and structural properties of this family of
genetic modifications.
We report herein the first incorporation into synthetic
oligonucleotides of a hydrophobic 1,N6-edAdo adduct
1, resulting from lipid peroxidation processes. The
adduct can be obtained in a high yield from a trans-2-
octenal epoxidized derivative generated in situ. The
strategy of protection developed to prepare the phos-
phoramidite derivative took into account the unstability
of the nucleoside adduct 1 under alkali conditions. After

V. M. Carvalho et al. / Bioorg. Med. Chem. 11 (2003) 2445–2452
2449
Experimental
Mass spectrometry measurements
Chemicals
Electrospray Ionization (ESI) mass spectrometry ana-
lyses of the adducts and derivatives together with those
of the modified oligonucleotides were conducted on
either a LCQ Ion Trap spectrometer (ThermoFinnigan,
San Jose, CA, USA) or a Platform 3000 spectrometer
(Micromass, Manchester, UK). Samples were diluted in
a solution of acetonitrile and water (50/50, v/v) that
contained either 0.1% formic acid (positive mode) or
1% triethylamine (negative mode) at a concentration of
10 mmol/L. The samples were introduced into the
electrospray ion source by a syringe pump at a flow rate
of 10 mL/min.
All reagents were of the highest available purity. Buffers
were prepared using water purified with a Milli-Q sys-
tem (Milford, MA, USA). Anhydrous solvents (di-
chloromethane, dimethylformamide, tetrahydrofuran,
acetonitrile and pyridine) were purchased from SDS
(Peypin, France) and stored over 3 A molecular sieves.
HPLC grade acetonitrile was obtained from EM Science
(Gibbstown, NJ, USA). Dichloromethane, methanol
and the silica gel (70–200 mm) used for the low-pressure
column chromatography were from SDS. 20-Deoxy-
adenosine, formic acid, silica gel 60 F-254 TLC and
preparative silica gel 60 F-254 TLC (2 mm) plates were
acquired from Merck (Darmstadt, Germany). Tri-
fluoroacetic acid (TFA), triethylamine (TEA) and di-
isopropylethylamine (DIEA) were from Acros (Geel,
Belgium). 4-Dimethylamine-pyridine was from Fluka
MALDI mass spectra were obtained with either a time-
of-flight mass spectrometer Voyager-DE (Perseptive
Biosystems, Framingham, MA, USA) or a time-of-flight
mass spectrometer Biflex (Bruker Daltonik, Bremen,
Germany) equipped with a 337 nm nitrogen laser and a
pulsed delay source extraction. For the matrix, a mix-
ture of 3-hydroxypicolinic acid and picolinic acid in a
4:1 (w/w) ratio was dissolved in a 50% acetonitrile
aqueous solution and a small amount of Dowex-50W
50X8-200 (Sigma) cation exchange resin. Then, 1 mL
of an aqueous solution of the sample was added to 1 mL of
the matrix and the resulting solution was stirred. The
sample was subsequently placed on top of the target
plate and allowed to dry by itself. The spectra were
calibrated with a 1 pmol/mL solution of myoglobin (m/z
16 952), using the same assay conditions as described
for the oligonucleotides.
(Buchs, Switzerland). Trans-2-octenal, 3-(trimethyl-
1
silyl)-1-propanesulfonic acid, D2O99,9%, Oxone
,
1,1,1-trifluoroacetone and chloroacetaldehyde were
from Aldrich (Milwaukee, WI, USA). Phenoxyacetyl-
dAdo and phenoxyacetyl-dGuo phosphoramidites
were purchased from Pharmacia Biotech (Uppsala,
Sweden). All other unmodified deoxyribonucleoside
phosphoramidites, DNA synthesis reagents as well as
CPG supports were from Glen Research (Sterling,
VA, USA). Calf intestinal alkaline phosphatase was
acquired from Amershan Pharmacia Biotech
(Uppsala, Sweden). Nuclease P1 (Penicillium citrium)
was from USB Corporation (Cleveland, OH, USA).
All other chemicals were from Sigma (St. Louis, MO,
USA).
Preparation of 1-[3-(2-deoxy-ꢀ-D-erythro-pentofuranosyl)-
3H-imidazo[2,1-i]purin-7-yl]-1-hexanol (1-hexanol-"dAdo)
(1). To an acetonitrile solution (70 mL) of trans-2-octe-
nal (7.614 g, 60 mmol) an aqueous Na2EDTA solution
(60 mL, of 0.4 mM) was added. The resulting homo-
genous solution was cooled down to 0–1 ꢀC, followed by
addition of trifluoroacetone (12 mL) via a precooled
syringe. To this solution, a mixture of sodium bicarbo-
nate (7.8 g, 98 mmol) and Oxone1 (36.9 g, 60 mmol) was
added in portions over a period of 1 h. After 3 h, the
reaction mixture was poured into water (150 mL) and
extracted with methylene chloride (2 ꢁ 50 mL). The
combined organic extracts were dried over anhydrous
magnesium sulfate and the solvent was subsequently
removed under a N2 flow. Then, the resulting residue
was diluted in acetonitrile (100 mL) and added to a
20-deoxyadenosine solution (3 g, 12 mmol) in 0.1 M pH
7.4 sodium phosphate buffer (100 mL) and the resulting
mixture was stirred for 120 h at 50 ꢀC. Unreacted alde-
hyde was removed by extraction with methylene chlo-
ride (2 ꢁ 50 mL). The aqueous solution was lyophilized
and the residue was dissolved in 250 mL of a solution of
acetonitrile and water (60/40, v/v). The resulting solu-
tion was filtered through 0.22 mm Durapore membranes
and purified by preparative RP-HPLC [System: a Prod-
igy C18 preparative column (Phenomenex, Torrance,
CA—250 ꢁ 21.2 mm i.d., 10 mm). The elution was per-
formed with 40% acetonitrile in water at a flow rate of
9 mL/min. The fluorescence detection was achieved at
the following wavelengths: 310 nm (excitation) and
HPLC apparatus
Preparative HPLC was conducted using a Shimadzu
system (Kyoto, Japan) with two LC-6AD, a Rheodyne
injector, a SPD-M10AV photodiode array and a
RF-10AXL fluorescence detector controlled by
a
SCL-10AVP module and Class-VP software. Semi-pre-
parative and analytical HPLC were performed on a
system composed of a Merck L-6200 pump, a Rheo-
dyne injector, a Waters 990 photodiode array and a
Merck F1000 fluorescence detector. Several liquid
chromatographic columns and gradients were used for
the purification steps and analytical experiments (vide
infra).
Nuclear magnetic resonance spectroscopy
1H NMR (300 MHz) and 13C NMR spectra were
recorded either with a Bruker Avance Series DPX-300
or a DRX-500 MHz apparatus (Rheinstetten, Ger-
many). 31P NMR (101 MHz) spectra were acquired on a
Varian INOVA 400 MHz spectrometer (Palo Alto, CA,
USA) calibrated with 85% H3PO4 as an external stan-
dard. Chemical shifts are expressed in parts per million
(ppm) from 2,20-dimethyl-2-silapentane-5-sulfonate (DSS)
when spectra were acquired in D2Oor DMSO- d6 or
from tetramethylsilane (TMS) when they were recorded
in CDCl3, methanol-d4 or acetone-d6.

2450
V. M. Carvalho et al. / Bioorg. Med. Chem. 11 (2003) 2445–2452
500 nm (emission). The emission wavelength was shifted
up to 90 nm from the maximum in order to attenuate
the detection. The latter HPLC purification yielded 1 as
a white solid (2.6 g, 58%). UV: lmax 232, 261, 268, 278
and 300 nm (pH ꢂ4); lmax 224 and 277 nm (pH <4). Rf
(CH2Cl2/CH3OH 80/20): 0.1; ESI-MS (positive mode—
cone voltage=15 V): m/z 398 (15) (M+Na)+, 376 (100)
(M+H)+, 282 (5) (BH+Na)+, 260 (35) BH+2 ; 1H
NMR (DMSO-d6, 500 MHz): d 9.18 (H-2, s, 1H), 8.52
(H-8, s, 1H), 7.41 (H-11, s, 1H,), 6.48 (H-10, t, 1H,
2H), 2.64–2.80 (H-20,H-200, m, 2H), 2.06 (H-14, m, 2H),
1.66 (H-15, m, 2H), 1.37 (H-16, m, 2H), 1.25 (H-17, m,
2H), 1.05–1.13 (iPr-TIPDS, m, 28H), 0.91 (H-18, t, 3H).
30,50-O-(1,1,3,3-Tetraisopropyldisiloxane-1,3-diyl)-[3-(2-
deoxy-ꢀ-D-erythro-pentofuranosyl)-3H-imidazo[2,1-i]-
purin-7-yl]-1-hexyl-2-methylpropanoate (3). Compound
2 (356 mg, 0.56 mmol) and 4-(dimethylamino)pyridine
(392 mg, 3.5 mmol) were coevaporated three times from
dry pyridine (5 mL). The mixture was redissolved in dry
pyridine (5 mL) and isobutyric anhydride (332 mL,
2 mmol) was added dropwise. After 30 min, the reaction
was completed as checked by TLC on silica gel plates
(80/20 CH2Cl2/CH3OH). The reaction was stopped by
the addition of methanol (2 mL) and the solvent was
removed under reduced pressure. A solution of 5%
NaHCO3 was added to the residue and the product was
extracted with dichloromethane (5 mL). Then, the
extract was dried over Na2SO4 and evaporated under
reduced pressure. Column chromatography on silica gel
(1–7.5% MeOH in CH2Cl2, as the eluent) yielded a
white powder product (331 mg, 83%); Rf (80/20
CH2Cl2/CH3OH): 0.77; ESI-MS (positive mode—cone
voltage=15 V): m/z 618 (15) (M+H)+, 260 (35) BH2+;
1H NMR (CDCl3, 300 MHz): d 8.95 (H-2, s, 1H), 8.16
(H-8, s, 1H), 7.63 (H-11, s, 1H,), 6.39 (H-10, t, 1H), 6.28
(H-13, m, 1H), 4.93 (H-30, m, 1H), 3.94 (H-40, m, 1H),
4.06 (H-50, H-500, m, 2H), 2.66–2.76 (H-20, H-200, m, 2H),
2.56 (CH-iBu, m, 1H) 2.14 (H-14, m, 2H), 1.41 (H-15,
m, 2H), 1.25–1.28 (H-16, H-17 m, 4H), 1.10–1.19 (CH3-
iBu, d, 6H), 1.04–1.10 (iPr-TIPDS, m, 28H), 0.88 (H-18,
t, 3H).
0
0
0
00
J1 ,2 =6.9 Hz, J1 ,2 =6.3 Hz), 5.48 (13-OH, d, 1H, exch.,
0
J13,13-OH=6.1 Hz), 5.34 (30-OH, d, 1H, exch., J3 ,3 -
0
OH=4.2 Hz), 5.01 (H-13, q, 1H, J13,14=6.6 Hz), 4.95 (50-
OH, t, 1H, J5 ,5 -OH=5.5 Hz), 4.43 (H-30, m, 1H,
0
0
J3 ,4 =3.0 Hz), 3.89 (H-40, m, 1H, J4 ,5 =3.7 Hz,
0
0
0
0
0
0
00
0
00
J4 ,5 =4.7 Hz), 3.60 (H-5 , m, 1H, J5 ,5 =ꢃ12.5 Hz),
3.55 (H-500, m, 1H), 2.73 (H-20, m, 1H, J2 ,2 =ꢃ14.0 Hz,
0
00
00
0
0
00
0
J2 ,3 =6.8 Hz), 2.36 (H-2 , m, 1H, J2 ,3 =3.9 Hz), 1.92
(H-14, td, 1H), 1.51 (H-15, m, 1H), 1.29 (H-16, m, 2H),
1.23 (H-17, m, 2H), 0.85 (H-18, t, 3H); 13C NMR
(DMSO-d6, 125 MHz): d 141.0 (C-6), 139.9 (C-8), 138.1
(C-4), 136.0 (C-2), 129.6 (C-11), 127.7 (C-12), 123.2 (C-
5), 88.1 (C-40), 84.1 (C-10), 70.8 (C-30), 63.6 (C-13), 61.8
(C-50), 40.0 (C-20), 34.9 (C-14), 31.2 (C-15), 25.3 (C-16),
22.2 (C-17), 14.1 (C-18).
30,50-O-(1,1,3,3-Tetraisopropyldisiloxane-1,3-diyl)-[3-(2-
deoxy-ꢀ-D-erythro-pentofuranosyl)-3H-imidazo[2,1-i]-
purin-7-yl]-1-hexanol (2). Compound
1
(261 mg,
0.70 mmol) was combined with imidazol (379 mg,
3.10 mmol) and the mixture was coevaporated twice in
dry pyridine. The residue was dissolved in dry dime-
thylformamide (4 mL) with triethylamine (197 mL,
1.4 mmol) and the resulting solution was cooled to 0 ꢀC
and kept under stirring. TIPDSCl2 (468 mL, 0.76 mmol)
was added in two portions over 1 h. Then, the mixture
was allowed to warm to room temperature while stirring
was maintained. After 2 h, TLC analysis showed that
less than 10% of the starting material was left. The
reaction was stopped by the addition of water (2 mL).
The mixture was neutralized by the addition of 5%
sodium bicarbonate solution (5 mL) and the products
were extracted with methylene chloride (2 ꢁ 10 mL).
The combined organic extracts were dried over anhy-
drous magnesium sulfate, evaporated to dryness and
coevaporated twice with anhydrous methylene chloride.
TLC on silica gel (80/20 CH2Cl2/CH3OH, as eluent)
showed the formation of one major product and a more
hydrophobic additional compound. Column chromato-
graphy on silica gel using 1–10% MeOH in CH2Cl2, as
the eluent, was performed for the purification of the
products. MS analysis (vide infra) showed that the main
product was the expected protected nucleoside whereas
the formation of the second product is explained by the
addition of two TIPDSCl2 groups. After combining and
evaporating the corresponding fractions, the pure
desired product 2 was obtained as a white powder
(350 mg, 70%). Rf (80/20 CH2Cl2/CH3OH): 0.58; ESI-
MS (positive mode—cone voltage=15 V): m/z 618 (15)
(M+H)+, 260 (35) BH2+; 1H NMR (CDCl3, 300 MHz):
d 9.08 (H-2, s, 1H), 8.15 (H-8, s, 1H), 7.40 (H-11, s,
1H,), 6.39 (H-10, t, 1H), 4.94 (H-13, m, 1H), 4.95, 4.93
(H-30, m, 1H) 3.94 (H-40, m, 1H), 4.53 (H-50, H-500, m,
1-[3-(2-Deoxy-ꢀ-D-erythro-pentofuranosyl)-3H-imidazo-
[2,1-i]purin-7-yl]-1-hexyl-2-methylpropanoate (4). Com-
pound 3 (234 mg, 0.34 mmol) was dried by coevapora-
tion from dry pyridine (3 ꢁ 2 mL) and redissolved in dry
dimethylformamide (5 mL). To this solution, CsF
(500 mg, 3.3 mmol) was added and the mixture was
stirred for 24 h. TLC analysis on silica gel plates with a
mixture of CH2Cl2/CH3OH (80/20, v/v) showed that the
starting product was fully consumed. The mixture was
poured into 5% aq NaHCO3 (10 mL), extracted with
dichloromethane (3 ꢁ 5 mL), and the organic solution
was dried over Na2SO4 and finally evaporated under
reduced pressure. The resulting residue was purified by
column chromatography on silica gel (1–15% MeOH in
CH2Cl2, as the eluent) yielding a white powder product
(129 mg, 85%). Rf (80/20 CH2Cl2/CH3OH): 0.32; ESI-
MS (positive mode—cone voltage=15 V): m/z 618 (15)
(M+H)+, 260 (35) BH2+; 1H NMR (CDCl3, 300 MHz):
d 9.21 (H-2, s, 1H), 8.48 (H-8, s, 1H), 7.59 (H-11, s, 1H),
6.65 (H-10, t, 1H), 5.25 (H-13, m, 1H), 4.64 (H-30, m,
1H), 4.21 (H-40, m, 1H), 3.81–3.87 (H-50, H-500, m, 2H),
2.98 (H-20, m, 1H), 2.65 (H-200, m, 1H), 2.39 (CH-iBu,
m, 1H), 2.13 (H-14, m, 2H), 1.55 (H-15, m, 2H), 1.29–
1.44 (H-16, H-17, m, 4H), 1.06 (CH3-iBu, d, 6H), 0.88
(H-18, t, 3H).
1-[3-(2-Deoxy-5-(4,40-dimethoxytrityl)-ꢀ-D-erythro-pento-
furanosyl)-3H-imidazo[2,1-i]purin-7-yl]-1-hexyl-2-methyl-
propanoate (5). Compound 4 (125 mg, 0.28 mmol) was
dried by coevaporation from dry pyridine (3 ꢁ 2 mL)

V. M. Carvalho et al. / Bioorg. Med. Chem. 11 (2003) 2445–2452
2451
and then redissolved in dry pyridine (5 mL). The solu-
tion was cooled in an ice/water bath and kept under an
argon atmosphere. Then, 4,40-dimethoxytrityl chloride
(DMTrCl, 108 mg, 0.32 mmol) was added under stir-
ring. The cooling bath was removed. After 1 h, it was
found that the reaction was completed as inferred from
TLC analyses on silica gel plates with a mixture of
CH2Cl2/CH3OH (80/20, v/v). The reaction was stopped
by the addition of methanol (1 mL) and the resulting
solution was evaporated to dryness. The oily residue
was dissolved in ethyl acetate (5 mL) and washed with
5% aq NaHCO3 (5 mL). The organic extract was dried
over Na2SO4 and evaporated under reduced pressure.
The resulting residue was purified by column chroma-
tography on silica gel (1–8% MeOH in CH2Cl2, as the
eluent) yielding a yellowish foam product (172 mg,
85%). Rf (80/20 CH2Cl2/CH3OH): 0.66; ESI-MS (posi-
tive mode—cone voltage=15 V): m/z 748.9 (100)
(M+H)+, 330.4 (35) BH2+; 1H NMR (CDCl3,
300 MHz): d 8.89 (H-2, s, 1H), 8.52 (H-8, s, 1H), 7.61 (H-
11, s, 1H,), 6.77–7.41 (aromatic H of DMTr, m, 13H),
6.52 (H-10, t, 1H), 6.26 (H-13, m, 1H), 4.43 (H-30, m, 1H),
3.89 (H-40, m, 1H), 3.76 and 3.77 (CH3–O–DMTr, s,
6H), 3.55–3.60 (H-50, H-500, m, 2H), 2.82 (H-20, m, 1H),
2.55 (H-200, m, 1H), 2.53 (CH–iBu, m, 1H), 2.13 (H-14,
m, 2H), 1.55 (H-15, m, 2H), 1.25–1.38 (H-16, H-17, m,
4H), 1.13 (CH3–iBu, d, 6H), 0.88 (H-18, t, 3H).
was placed in 0.5 mL of the following solutions: 32%
aqueous ammonia, 80% acetic acid aqueous solution,
0.1 M oxidizing solution of iodine and 0.05 M K2CO3
methanolic solution. The solutions were kept at room
temperature and the reactions were stopped at increas-
ing time intervals (0, 1, 2, 4, 8 and 24 h) by freezing the
samples in liquid nitrogen with subsequent lyophiliza-
tion. Then, samples were analyzed by HPLC [system: a
Luna C18 analytical column (Phenomenex—150 ꢁ
4.6 mm i.d., 3 mm) eluted with a linear binary gradient of
acetonitrile in 10 mM pH 7.0 TEAA ranging acetonitrile
concentration from 2 to 10% in 15 min, then from 10 to
45% in 15 min at a flow rate of 1 mL/min].
Solid-phase synthesis of oligonucleotides
The synthesis of all oligonucleotides was performed at a
1 mmol scale using the phosphoramidite chemistry on an
Applied Biosystems 392 automated Nucleic Acids syn-
thesizer. The Pac-protection system18 for the nucleobase
amino functions was used with regard to the high labi-
lity of 1 under alkali deprotection conditions. The
synthesis was performed on controlled pore glass (CPG)
supports following the standard protocol, at the excep-
tion of two modifications. First, the duration of cou-
pling of the modified nucleoside phosphoramidite was
extended to 10 min, instead of 30 s for normal nucleo-
sides. Under the latter conditions, an adduct coupling
efficiency up to 90% was achieved. Moreover, the cap-
ping step was carried out with a solution of phenox-
yacetic anhydride in THF to prevent possible
transacylation reactions.23 Finally, the 50-terminal
DMTr group was kept at the end of the synthesis (trityl-
on mode).
1-{2-Deoxy-3-O-[2-cyanoethoxy(diisopropylamino)-phos-
phino]-5-(4,40-dimethoxytrityl)-ꢀ-D-erythro-pentofurano-
syl)-3H-imidazo[2,1-i]purin-7-yl]-1-hexyl-2-methylpro-
panoate (6). Compound 5 (157 mg, 0.21 mmol) was
dried by coevaporation from dry pyridine (3 ꢁ 2 mL)
and then dissolved in dry dichloromethane (2.5 mL).
The solution was cooled in an ice/water bath and kept
under argon atmosphere. Diisopropylethylamine
(88 mL, 0.50 mmol) and 2-cyanoethyl N,N-diisopropyl-
chlorophosphoramidite (60 mL, 0.27 mmol) were added
to the solution under stirring at room temperature. The
course of the reaction was monitored by TLC with a
mixture of CH2Cl2/CH3OH/TEA (89/10/1, v/v/v). After
1 h, a total consumption of 5 was observed. The reac-
tion was then stopped by the addition of methanol
(2 mL). The mixture was diluted with ethyl acetate
(5 mL) and subsequently washed with 5% aq NaHCO3
(5 mL). The organic layer was dried over Na2SO4 and
evaporated under reduced pressure. Then, the products
were purified by RP-HPLC [System: a Bondclone C18
semi-preparative column (Phenomenex—300 ꢁ 7.8 mm
i.d., 10 mm) eluted with 20% acetonitrile in 10 mM
TEAA, pH 7.0, at a flow rate of 5 mL/min]. The col-
lected fractions were lyophilized yielding 6 as a light yel-
lowish foam (123 mg, 62%). The latter product was kept
under argon atmosphere at ꢃ20 ꢀC. Rf (89/10/1 CH2Cl2/
CH3OH/TEA): 0.77; ESI-MS (positive mode—cone
voltage=15 V): m/z 948 (100) (M+H)+, 330 (10) BH2+,
Deprotection and purification of oligonucleotides
The solid supports were treated with 50 mM K2CO3 in
anhydrous methanol (1.5 mL). After 12 h, the metha-
nolic solution was transferred to another tube and the
resin was washed with water (2 ꢁ 1 mL). The combined
extracts were immediately and carefully neutralized with
acetic acid. The solvent was removed under reduced
pressure and the residue was dissolved in water
(200 mL). The crude synthetic mixture was purified and
detritylated on-line by RP-HPLC following a protocol
previously described.20 Finally, the 14- and 22-mer oligo-
nucleotides were further purified by preparative 20%
polyacrylamide gel electrophoresis, to afford 24 AU260nm
and 31 AU260nm, respectively.
Enzymatic digestion of the modified oligonucleotides by
nuclease P1, calf spleen phosphodiesterase, bovine
intestinal mucosa phosphodiesterase and alkaline
phosphatase
303 (5) (trityl)+
;
149.31 and 149.20 ppm (two diastereoisomers).
31P NMR (acetone-d6, 101.21 MHz): d
The modified oligonucleotides (0.4 UA260nm) were taken
up in 45 mL of water and completely digested into
nucleotides upon incubation with 3 mL of a solution of
nuclease P1 (1 U/mL) and 1 mL of calf spleen phospho-
diesterase (0.004 U) in acetate buffer (0.3 M sodium
acetate/1 mM ZnCl2, pH=5.3) for 2 h at 37 ꢀC. Then,
5 mL of Tris buffer (1 M Tris–HCl, 2 mM EDTA,
Stability studies of 1-[3-(2-deoxy-ꢀ-D-erythro-pentofura-
nosyl)-3H-imidazo[2,1-i]purin-7-yl]-1-hexanol (1) under
the alkaline, acid and oxidizing conditions used for che-
mical synthesis of oligonucleotides. 30 mg of compound 1

2452
V. M. Carvalho et al. / Bioorg. Med. Chem. 11 (2003) 2445–2452
pH=8.5) together with 2 units of alkaline phosphatase
and 0.003 U of bovine intestinal mucosa phospho-
diesterase were added and the resulting solution was
further incubated for 1 h at 37 ꢀC. The resulting diges-
tion mixture of 20-deoxyribonucleosides was centrifuged
and analyzed by RP-HPLC [system: a Hypersil 5 mm
C18 column (4.6 ꢁ 250 mm) eluted with TEAA (10 mM,
pH 7) and acetonitrile (linear gradient from 0 to 18% of
acetonitrile in 40 min at a flow rate of 1 mL/min)].
5. Nair, J.; Barbin, A.; Guichard, Y.; Bartsch, H. Carcino-
genesis 1995, 16, 613.
6. Marnett, L. J. DNA Adducts of a,b-Unsaturated Alde-
hydes and Dicarbonyl Compounds. In DNA Adducts: Identi-
fication and Biological Significance. IARC Scientific
Publications no. 125, Hemminki, K., Dipple, A., Shuker,
D. E. G., Kadlubar, F. F., Segerback, D., Bartsch, H. Eds.;
¨
IARC: Lyon, France, 1994; p 151.
7. el Ghissassi, F.; Barbin, A.; Nair, J.; Bartsch, H. Chem.
Res. Toxicol. 1995, 8, 278.
8. Sodum, R. S.; Chung, F. L. Cancer Res. 1988, 48, 320.
9. Sodum, R. S.; Chung, F. L. Cancer Res. 1991, 51, 137.
10. Carvalho, V. M.; Di Mascio, P.; de Arruda Campos, I.;
Douki, T.; Cadet, J.; Medeiros, M. H. Chem. Res. Toxicol.
1998, 11, 1042.
11. Carvalho, V. M.; Asahara, F.; Di Mascio, P.; de Arruda
Campos, I.; Cadet, J.; Medeiros, M. H. Chem. Res. Toxicol.
2000, 13, 397.
12. Rindgen, D.; Lee, S. H.; Nakajima, M.; Blair, I. A. Chem.
Res. Toxicol. 2000, 13, 846.
13. Yang, D.; Wong, M. K.; Yip, Y. C. J. Org. Chem. 1995,
60, 3887.
14. Kuijpers, W. H.; Kuyl-Yeheskiely, E.; van Boom, J. H.;
van Boeckel, C. A. Nucleic Acids Res. 1993, 21, 3493.
15. Reddy, G. R.; Marnett, L. J. J. Am. Chem. Soc. 1995, 117,
5007.
16. Schnetz-Boutaud, N. C.; Mao, H.; Stone, M. P.; Marnett,
L. J. Chem. Res. Toxicol. 2000, 13, 90.
17. Chenna, A.; Singer, B. Chem. Res. Toxicol. 1997, 10,
165.
18. Schulof, J. C.; Molko, D.; Teoule, R. Nucleic Acids Res.
1987, 15, 397.
Acknowledgements
The authors thank Colette Lebrun, Christine Saint-
Pierre (SCIB/CEA-Grenoble) and Dr. Michel Jaquinod
(LSMP, IBS-Grenoble) for their contributions to the
mass spectrometry measurements. The assistance of
Pierre-Alain Bayle (SCIB/CEA-Grenoble) and Miriam
Uemi (IQ-USP, Sao Paulo) for the NMR analyses is
gratefully acknowledged. This work was supported by
the Fundacao de Amparo a Pesquisa do Estado de Sao
Paulo, FAPESP (Brazil), the Comite de Radio-
protection (Electricite de France), the Conselho Nacio-
nal para o Desenvolvimento Cientıfico e Tecnologico,
CNPq (Brazil) and the Comite Francais d’Evaluation de
la Cooperation Universitaire avec le Bresil (USP-
COFECUB/UC-23/96). V.M.C. was a recipient of a
FAPESP fellowship.
19. Markiewicz, W. T. J. Chem. Res. (S) 1979, 24.
20. Romieu, A.; Gasparutto, D.; Molko, D.; Cadet, J. Tetra-
hedron Lett. 1997, 38, 7531.
References and Notes
1. Kochetkov, N. K.; Shibaev, V. N.; Kost, A. A. Tetrahedron
Lett. 1971, 15, 1993.
2. Barbin, A. Mutat. Res. 2000, 462, 55.
3. Moriya, M.; Zhang, W.; Johnson, F.; Grollman, A. P.
Proc. Natl. Acad. Sci. U.S.A. 1994, 91, 11899.
4. Pandya, G. A.; Moriya, M. Biochemistry 1996, 35, 11487.
21. Butenandt, J.; Burgdorf, L. T.; Carell, T. Synthesis 1999,
7, 1085.
22. Gasparutto, D.; Bourdat, A. G.; D’Ham, C.; Duarte, V.;
Romieu, A.; Cadet, J. Biochimie 2000, 82, 19.
23. Chaix, C.; Molko, D.; Teoule, R. Tetrahedron Lett. 1989,
30, 71.