1H-benzotriazole as synthetic auxiliary in a facile route to N6- (arylmethyl)-2'-deoxyadenosines: DNA intercalators inserted into three-way junctions

Article abstract of DOI:10.1002/1522-2675(20000705)83:7<1408::AID-HLCA1408>3.0.CO;2-U

The 2'-deoxy-N⁶-(naphthalen-1-ylmethyl)- (5a) and N⁶-(pyren-1- ylmethyl)adenosine (5b) were synthesized in two steps from 2'-deoxyadenosine and the adequate arenecarbaldehyde with 1H-benzotriazole as a synthetic auxiliary (Scheme). W

Full text of DOI:10.1002/1522-2675(20000705)83:7<1408::AID-HLCA1408>3.0.CO;2-U

1408
Helvetica Chimica Acta ± Vol. 83(2000)
1H-Benzotriazole as Synthetic Auxiliary in a Facile Route to
N6-(Arylmethyl)-2'-deoxyadenosines: DNA Intercalators Inserted into
Three-Way Junctions
by Sherif A. El-Kafrawya), Magdy A. Zahranb), Erik B. Pedersena)*, and Claus Nielsenc)
a) Department of Chemistry, University of Southern Denmark, Odense Universitet, DK-5230 Odense M
b) Chemistry Department, Faculty of Sciences, Menoufia University, Egypt
c) Department of Virology, State Serum Institute, DK-2300 Copenhagen S
(fax: ‡ 4566158780, e-mail: EBP@Chem.sdu.dk)
Dedicated to Prof. Dr. Frank Seela on the occasion of his 60th birthday
The 2'-deoxy-N6-(naphthalen-1-ylmethyl)- (5a) and N6-(pyren-1-ylmethyl)adenosine (5b) were synthe-
sized in two steps from 2'-deoxyadenosine and the adequate arenecarbaldehyde with 1H-benzotriazole as a
synthetic auxiliary (Scheme). When the N6-(arylmethyl)-2'-deoxyadenosines were inserted into the junction
region of a DNA three-way junction, its thermal stability increased.
1. Introduction. ± Carcinogenic polycyclic aromatic hydrocarbons are known to
undergo metabolic activation to reactive diol epoxide intermediates that bind
covalently to DNA in vivo [1 ± 3]. Moreover, the presence of bulky polycyclic aromatic
residues derived from the binding of stereoisomeric mutagenic and tumorigenic
metabolites of benzo[a]pyrene to 2'-deoxyguanosine residues in DNA is known to
block replication [4] and to induce mutations [5], and is believed to constitute the
critical initial step in carcinogenesis [6]. To gain greater insight into polyaromatic
hydrocarbon (PAH) adducts to DNA and the biological consequences of their
replication, efficient methods for the site-specific synthesis of PAH-oligodeoxynucleo-
tides, with the PAH attached covalently to the exocyclic amino groups of specific dA
and dG bases, are required.
Oligonucleotides complementary to strategic regions of viral or messenger RNAs
were first shown by Zamecnik and Stephenson [7] to inhibit viral replication. The
factors that are usually assumed to limit the activity of antisense oligonucleotides are
cellular uptake, resistance to nuclease, and the stability of the hybrid formed.
Modifications are usually chosen to improve one or more of these properties. However,
a steady improvement in activity has been achieved by this rational approach that is
encouraging for the design of conjugates to meet specific requirements. The
manipulation of sequence-specific protein binding and gene expression with antigene
or antisense technology requires oligonucleotides that bind DNA or RNA with high
affinity and specificity [8 ± 11]. Oligonucleotides capable of selective recognition of
RNA must accommodate, in addition, both the kinetic [12][13] and thermodynamic
[13][14] consequences of nonuniform RNA conformation. The structural complexity
inherent in RNA complicates recognition because it limits the accessibility of single-
stranded regions that must be differentiated for specific binding. It is estimated that

Helvetica Chimica Acta ± Vol. 83(2000)
1409
11 ± 15 nucleotides are necessary to define a unique mRNA sequence in eukaryotic cell
[15], yet few well-characterized RNA secondary structures contain contiguous single-
stranded regions of this size.
The object of this investigation will be focused on the effect of intercalators as a
chemical modification on strengthening the binding force in an oligonucleotide three-
way junction (TWJ). It is believed that three-way junctions with two short single-
stranded arms are interesting target sites for antisense oligonucleotides when
contiguous single-stranded regions are not available for binding. However, the stability
of the hybridization of oligodeoxynucleotides to TWJ single-stranded arms is lower
compared to the corresponding duplex with the same number of base pairs [16][17].
2. Synthesis. ± In continuation of our previous work [18], we now report a facile and
short direct route to synthesize 2'-deoxy-N6-(naphthalen-1-ylmethyl)adenosine (5a)
and 2'-deoxy-N6-(pyren-1-ylmethyl)adenosine (5b) by means of 1H-benzotriazole [19]
as a facile and versatile reagent, and this in good yield and without any by-products
[20][21] arising from direct alkylation at other sites, e.g., at N(1), N(3 ) or N(7).Harvey
and co-workers [22] reported recently the synthesis of 2'-deoxy-N6-(pyren-1-ylmethyl)-
adenosine (5b), which involves in the key step coupling of an appropriately protected
halopurine derivative with the amino derivative of the polyaromatic hydrocarbon, in
this case pyrene-1-methanamine. Our synthetic strategy entailed in the key step the
condensation of 1H-benzotriazole (2) with the appropriate polyaromatic aldehydes
3a,b and with 2'-deoxyadenosine (1) by refluxing this mixture in toluene with
azeotropic removal of H2O (Dean-Stark) in an acid-catalyzed reaction, i.e., the reaction
took place in the presence of a few drops of AcOH (Scheme). After ca. 6 h, the solvent
was evaporated, and the 1H-benzotriazole adducts 4a or 4b were obtained from the
residue after flash-chromatographic (FC) purification in 85 and 80% yield, respec-
tively. The 1H-benzotriazole adducts 4a and 4b were then reduced by LiAlH4 in dry
THF to give 5a or 5b in 80 and 70% yield, respectively, after FC purification. The struc-
tures of the adducts 4a,b were elucidated from their 1H- and 13C-NMR and FAB mass
spectra. The former spectra were complex due to the presence of the stereogenic C-atom
Bt-CH-Ar. The diastereomer mixtures could not be separated by column chromatography.
In our previous study [18], we had the problem of very bad yields, or no reaction at
all, when the reaction between 1H-benzotriazole (2), 2'-deoxyadenosine (1), and
polycyclic aromatic aldehyde 3 was carried out by reflux in EtOH in the presence of a
catalytic amount of AcOH. We believed that this may be due to the steric hindrance
around the stereogenic C-atom (N6-CH) generated by the presence of the 1H-
benzotriazol-1-yl and the bulky polycyclic aromatic group at the same C-atom. We also
thought that the H2O liberated during the condensation reaction could retard the
reaction progress. Therefore, we now modified the reaction conditions by carrying out
the reaction in refluxing toluene, using the Dean-Stark trap to remove azeotropically
the H2O formed, to force the reaction to proceed to completion.
The reduction of the adducts 4a,b was performed with LiAlH4 in THF at room
temperature to give the N6-(arylmethyl)deoxyadenosines 5a,b in 70 ± 80% yield.
Attempts to carry out the reduction with NaBH4 resulted in poor yields and required
long reaction times. The spectra of 2'-deoxy-N6-(pyren-1-ylmethyl)adenosine (5b)
were in good agreement with those reported [24].

1410
Helvetica Chimica Acta ± Vol. 83(2000)
Scheme
Naph ˆ naphthalen-1-yl, Pyrˆ pyren-1-yl, Bt ˆ 1H-benzotriazol-1-yl, DMTˆ 4,4'-dimethoxytrityl ˆ (MeO)2Tr
The 4,4'-dimethoxytrityl ((MeO)2Tr), and the phosphoramidite derivatives of 5a,b,
i.e., 6a,b and 7a,b, respectively, were prepared according to Caruthersꢀ methodology
[23]. For the preparation of 7a,b the (MeO)2Tr derivatives 6a,b were treated with
2-cyanoethyl diisopropylphosphoramidochlorite in the presence of iPr2EtN. The purity
of the obtained phosphoramidites 7a,b (yield 83and 85%, resp.) was 94 and 99%,
respectively, according to 31P-NMR, the remaining 1 ± 6% being non-phosphoramidite
by-products.
The modified and unmodified oligodeoxynucleotides (ODNs) were synthesized
on a Pharmacia Gene-Assembler-Special synthesizer on a 0.2-mmol scale according to
the standard phosphoramidite method [23] with 7 and commercial phosphorami-
dites. Phosphorothioates were prepared with the Beaucage reagent [25][26] (3H-1,2-
benzodithiol-3-one 1,1-dioxide) on a 0.2-mmol scale. The ability of ODNs to hybridize
to their complementary DNA strand was examined by UV melting measurements.
The melting points Tm were determined as the maxima of the first derivative of the
melting curves. DNA Three-way junctions were formed from equimolar amounts
(3 mm) in each strand at pH 7.0 in a 1 mm EDTA, 10 mm Na2HPO4, and 140 mm NaCl
buffer.

Helvetica Chimica Acta ± Vol. 83(2000)
1411
3. Stability of Three-Way Junction. ± To study the intercalating ability of the
prepared modified nucleosides 5, we chose a three-way junction composed of two
dangling single strands and a double-stranded DNA stem rich in C ´ G base pairs to shift
the melting temperature of the stem to a higher temperature to avoid its interference
with those obtained when hybridizing the dangling strands (Fig. 1).
Fig. 1. Three-way junction I. XX ˆ Site of insertion. The nucleoside symbols represent 2'-deoxynucleosides.
On inserting compound 5a and an extra dG as a bulge in the junction region of the
three-way junction I (Table 1, Entry 3), an increase in Tm of 6.48 was observed
compared to the insertion of dA and dG as a reference (Entry 2). Insertion of 5a in
both the 5'-end and the junction region had a small further effect (Entry 7, DTm ˆ 6.88)
when compared to the reference with dA instead of 5a (Entry 4). When the reference
ODN had 5a added at the 5'-end and dG and dA in the junction region (Entry 6), a
DTm ˆ 3.28 was observed when 5a was inserted instead of dA in the junction region
(Entry 7). Mismatches in the targeting ODN resulted in remarkable decreases in the
Table 1. Hybridization Data for the Three-Way Junction I with Inserted dG. X ˆ 5a.
Entry
Targeting ODN
Tm [8C]
1
2
3
4
5
6
7
8
9
5'-d(CTGTACCGT
5'-d(CTGTACCGT
5'-d(CTGTACCGT
5'-d(ACTGTACCGT AG CGCCCA)-3'
5'-d(ACTGTACCGT XG CGCCCA)-3'
5'-d(XCTGTACCGT AG CGCCCA)-3'
5'-d(XCTGTACCGT XG CGCCCA)-3'
CGCCCA)-3'
AG CGCCCA)-3'
XG CGCCCA)-3'
41.2
44.8
51.2
44.0
48.4
47.6
50.8
3
5'-d(CTTGACCGT
5'-d(CTGTACCGT
XG CGCCCA)-3'a)
XG CCGCCA)-3'a)
1.2
40.0
a) Mismatch site in the targeting ODN is underlined.

1412
Helvetica Chimica Acta ± Vol. 83(2000)
stabilization of the three-way junction (Entries 8 and 9). This confirms hybridization in
both arms of TWJ.
Insertion of a dG in the junction region could lead to base pairing with the first dC
in the stem, leading to the disruption of the corresponding first base pair in the stem. In
fact, DNA folding calculations [27] show that inserted d(AG) in the junction region
leads to disruption of the first two base pairs in the stem (d(GC) and d(AT)). To
overcome this problem, another antisense ODN containing dT and dA and dT and 5a
insertions in the junction region was used. The hybridization data shown in Table 2
showed a DTm ˆ 6.08 when inserting dT and 5a in the junction region (Entry 11)
compared to the reference with d(TA) in the same region (Entry 13). A more
pronounced effect (DTm ˆ 13.68) was achieved when two modified nucleosides 5a
(Entry 12) were inserted in the junction region compared to the insertion of two dA
(Entry 14) in the same region. As a preliminary test for the increased binding
properties of 5b over 5a, a single trial was made with the insertion of dT and 5b in the
junction region, and a DTm ˆ 10.48 was observed (Entry 11 vs. 13). We have recently
suggested that intercalators inserted at TWJ junctions stabilize the TWJ by stacking on
top of the deflecting arm [28]. Further stabilization by insertion of two intercalators
(Entry 12) could then be due to stacking of both intercalators at the top of the
deflecting arm.
Table 2. Hybridization Data for the Three-Way Junction I Avoiding Inserted dG. X ˆ 5a.
Entry
Targeting ODN
Tm [8C]
10
11
12
13
14
5'-d(CTGTACCGT
CGCCCA)-3'
41.2
5'-d(CTGTACCGT TX CGCCCA)-3'
5'-d(CTGTACCGT XX CGCCCA)-3'
5'-d(CTGTACCGT TA CGCCCA)-3'
5'-d(CTGTACCGT AA CGCCCA)-3'
44.8 (49.2) a)
51.2
38.8
37.6
a) Value in parentheses for X ˆ 5b.
The human immunodeficiency virus (HIV) contains at least two structured RNAs
that interact with viral protein and mediate novel genetic regularity pathways. One of
these is a 234-nucleotide sequence located within the env coding region called the Rev
response element (RRE). Interaction of the RRE with viral protein Rev [29][30]
regulates the appearance of unspliced or singly spliced viral mRNAs in the cytoplasm
of infected cells [31 ± 40]. Cload and Schepartz [41] studied the in vitro inhibitory effect
of some tethered oligonucleotide probes (TOPs) on the function of RRE. We chose the
same antisense sequence (Fig. 2) as that reported by Cload and Schepartz to have
potent concentration-dependent inhibition of RRE function in vitro. Only the tethering
propane-1,3-diol moiety was replaced with an intercalating nucleoside.
Thus, a DTm of 12.28 was observed when inserting 5b and dG in the junction region
(Table 3, Entry 16) compared to Entry 15. Also mismatching by interchanging two
nucleosides in the antisense ODN decreased the stability of the system (Entries 17 and
18), indicating that full complementarity took place between the three strands of the
system.

Helvetica Chimica Acta ± Vol. 83(2000)
1413
Fig. 2. Three-way junction II. XG ˆ Site of insertion. The nucleoside symbols represent 2'-deoxynucleosides.
Considering the stabilizing effect of the pyrenyl intercalator on DNA TWJ, it was
found interesting to test the activity of the corresponding phosphorothioates against
HIV-1 in MT-4 cells. The ODNs given in Table 3 were also synthesized as their
phosphorothioate analogues, as well as additional examples with dA, d(AA), or d(XX)
insertions in the middle of the oligonucleotide. The HIV-1 strain HTLV-IIIB was
propagated in MT-4 cells at 378 with RPMI-1640, 10% heat-inactivated fetal-calf serum
(FCS), and antibiotics (growth medium). The cells were preincubated with lipofectin
to improve their uptake of the oligonucleotide. Unfortunately, none of the prepared
phosphorothioates showed any anti-HIV activity. Moreover, they were cytotoxic at
100 mm.
Table 3. Hybridization Data for the Three-Way Junction II. X ˆ 5b.
Entry
Targeting ODN
Tm [8C]
15
16
17
18
5'-d(CTGTACCG
CGCCCA)-3'
40.8
53.0
33.2
40.4
5'-d(CTGTACCG XG CGCCCA)-3'
5'-d(CTGTACCG XG CCGCCA)-3' (m) a)
5'-d(CTTGACCG XG CGCCCA)-3' (m) a)
a) Mismatch site in the targeting ODN is underlined.
Experimental Part
General. Anal. TLC: Merck precoated silica gel 60 F254 plates. Flash column chromatography (FC): silica
gel (0.040 ± 0.063mm) from Merck. NMR Spectra: at 300 (1H), 75.5 (13C), and 121.5 MHz (31P); Varian-Gemini
2000 300-MHz spectrometer; d values in ppm rel. to SiMe4 as an internal standard (1H, 13C) and rel. to 85%
H3PO4 as external standard (31P). FAB-MS: positive mode; Kratos MS-50-RF spectrometer.
Oligonucleotides. Melting experiments were carried out on a Perkin-Elmer UV/VIS spectrometer Lambda
2 fitted with a PTP-6-Peltier temp.-programming element. The UV absorbance at 260 nm as a function of time
was recorded while the temp. was raised gradually (18/min) in a 1-cm cuvette, at a 3 mm concentration of each
component. DNA Syntheses were performed on a Pharmacia Gene-Assembler-Special DNA synthesizer, on a

1414
Helvetica Chimica Acta ± Vol. 83(2000)
0.2-mmol scale. The coupling efficiencies for the modified amidites (20-min couplings) 7a,b were ca. 85%
compared to 99% for the commercial ones (2-min coupling). The efficiency of each coupling step was monitored
by the release of dimethoxytrityl cation after each step. Removal from solid support and deprotection was
carried out at r.t. for 4 days in 25% ammonia and 0.01m EDTA in 25% ammonia for the phosphorothioates.
Desalting of the oligonuclotides was accomplished with disposable NAP-10 columns (Pharmacia).
N6-[(1H-Benzotriazol-1-yl)(naphthalen-1-yl)methyl]-2'-deoxyadenosine (4a). A soln. of 2'-deoxyadeno-
sine (1; 2.51 g, 10 mmol), 1H-benzotriazole (2; 1.43g, 12 mmol), and naphthalene-1-carboxaldehyde ( 3a;
1.6 ml, 12 mmol) was refluxed in toluene in the presence of 5 drops of AcOH as a catalyst with azeotropic
removal of H2O in a Dean-Stark apparatus. After 6 h, the solvent was evaporated and the residue submitted to
‡
FC (CHCl3/MeOH 99 :1): 4a (4.3g, 85%). FAB-MS (CHCl 3 ‡ 3-nitrobenzyl alcohol): 509 ([M ‡ H] ).
N6-[(1-H-Benzotriazol-1-yl)pyren-1-yl)methyl]-2'-deoxyadenosine (4b): As described for 4a, with 1 (2.51 g,
10 mmol), 2 (1.43g, 12 mmol), pyrene-1-carboxaldehyde ( 3b; 2.76 g, 12 mmol), toluene and 5 drops of AcOH:
‡
4b (4.6 g, 80%). FAB-MS (CHCl3 ‡ 3-nitrobenzyl alcohol): 583 ([M ‡ H] ).
2'-Deoxy-N6-(naphthalen-1-ylmethyl)adenosine (5a). A soln. of 4a (2.55 g, 5 mmol) was stirred with a
suspension of LiAlH4 (1.14 g, 30 mmol) in dry THF at r.t. for 2 h. The mixture was then poured into ice-water,
neutralized with AcOH and extracted with CHCl3. The crude mixture was purified by FC (CHCl3/MeOH 97:3):
5a (1.56 g, 80%). 1H-NMR (CDCl3): 8.39 (s, HÀC(8)); 8.06 (s, HÀC(2)); 7.86 ± 7.36 (m, Naph.); 6.12 (m,
HÀC(1')); 5.25 (br., NHCH2Ar); 4.69 (d, 1 OH); 4.07 (m, HÀC(3')); 3.92 (m, HÀC(4')); 3.77 (m, 2HÀC(5'));
2.94 (m, 1HÀC(2')); 2.17 (m, 1 HÀC(2')). 13C-NMR (CDCl3): 154.86 (C(6)); 152.71 (C(2)); 147.86 (C(4));
139.47 (C(8)); 133.88, 133.36, 131.53, 128.83, 128.70, 126.63, 126.49, 126.03, 125.42, 123.48 (Naph); 121.17
(C(5)); 89.54 (C(4'); 87.56 (C(1')); 73.06 (C(3')); 63.25 (C(5')); 42.57 (NHCH2Ar); 40.67 (C(2')). FAB-MS
‡
(CHCl3 ‡ 3-nitrobenzyl alcohol): 392 ([M ‡ H] ).
2'-Deoxy-N6-(pyren-1-ylmethyl)adenosine (5b). As described for 5a, with 4b (1.16 g, 2 mmol) and LiAlH4
(0.456 g, 12 mmol) in dry THF: 5b (0.65 g, 70%). 1H-NMR ((D6)DMSO): 8.60 (s, HÀ(C8)); 8.56 (s, HÀC(2));
8.42 ± 8.05 (m, Pyr); 6.41 (m, HÀC(1')); 5.49 (br., NHCH2Ar); 5.35 (d, 1 OH); 4.61 (t, 1 OH); 4.46 (m,
HÀC(3')); 3.94 (m, HÀC(4')); 3.61 (m, 2 HÀC(5')); 2.78 (m, 1 HÀC(2')); 2.32 (m, 1 HÀC(2')). 13C-NMR
(CDCl3): 154.30 (C(6)); 152.36 (C(2)); 147.79 (C(4)); 139.03 (C(8)); 131.19, 130.51, 130.15, 130.00, 128.10,
127.23, 126.76, 126.62, 125.93, 25.41, 124.62, 124.55, 124.10, 123.90, 122.32 (Pyr); 120.48 (C(5)); 88.74 (C(4'));
86.33 (C(1')); 71.61 (C(3')); 62.50 (C(5')); 41.58 (NHCH2Ar), 40.03(C(2 ')). FAB-MS (CHCl3 ‡ 3-nitrobenzyl
‡
alcohol): 466 ([M ‡ H] ).
4,4'-Dimethoxytrityl Derivatives 6: General Procedure. Compound 5a or 5b (2 mmol) was stirred in pyridine
at r.t. with 2.2 mmol of 4,4'-dimethoxytrityl chloride for 3h. The pyridine was evaporated and the residue
submitted to FC (CHCl3): yield 90%.
2'-Deoxy-5'-O-(4,4'-dimethoxytrityl)-N6-(naphthalen-1-ylmethyl)adenosine (6a): 1H-NMR (CDCl3): 8.42 (s,
HÀC(8)); 8.06 (s, HÀC(2)); 7.84 ± 7.13( m, arom. H); 6.42 (m, HÀC(1'), NH); 5.30 (br., NHCH2Ar); 4.62 (m,
OH); 4.10 (m, HÀC(3')); 3.80 ± 3.69 (m, MeO, HÀC(4')); 3.37 (m, 2 HÀC(5')); 2.72 (m, 1 HÀC(2')); 2.47 (m, 1
HÀC(2')). 13C-NMR (CDCl3): 154.95 (C(6)); 152.25 (C(2)); 149.63 (C(4)); 138.25 (C(8)); 136.37 ± 123.52
(arom. C), 120.04 (C(5)); 113.20 ((MeO)2Tr); 86.01 (C(4')); 84.20 (C(1')); 72.22 (C(3')); 63.64 (C(5')); 55.11
‡
(MeO); 42.80 (NHCH2); 40.22 (C(2')). FAB-MS (CHCl3 ‡ 3-nitrobenzyl alcohol): 694 ([M ‡ H] ).
2'-Deoxy-5'-O-(4,4'-dimethoxytrityl)-N6-(pyren-1-ylmethyl)adenosine (6b): 1H-NMR (CDCl3): 8.33 (s,
HÀC(8)); 8.30 (s, HÀC(2)); 8.17 ± 7.95 (m, Pyr); 7.37 ± 7.16 (m, (MeO)2Tr); 6.25 (t, HÀC(1')); 5.53(br.,
NHCH2Ar), 4.55 (m, OH); 4.03( m, HÀC(3')); 3.70 (s, MeO); 3.30 (m, HÀC(5')); 2.60 (m, 1 HÀC(2')); 2.33
(m, 1 HÀC(2')). 13C-NMR (CDCl3): 154.45 (C(6)); 153.23 (C(2)); 144.59 (C(4)); 138.13 (C(8)); 135.72 ± 122.93
(arom. C); 120.08 (C(5)); 113.16 ((MeO)2Tr); 85.47 (C(4')); 84.02 (C(1')); 72.19 (C(3')); 65.80 (C(5')); 55.07
‡
(MeO); 42.82 (NHCH2Ar); 40.00 (C(2')). FAB-MS (CHCl3 ‡ 3-nitrobenzyl alcohol): 768 ([M ‡ H] ).
Phosphoramidite Derivatives 7: General Procedure. Compound 6a or 6b (0.5 mmol) was co-evaporated with
dry MeCN and dissolved in a mixture of iPr2EtN (0.4 ml) and dry CH2Cl2 (2.5 ml). Then, 2-cyanoethyl
diisopropylphosphoramidochlorite (0.8 mmol) was added dropwise under Ar. The mixture was stirred at r.t. for
1 h. Then the reaction was quenched by addition of MeOH (0.2 ml). After addition of AcOEt and CH2Cl2, the
org. phase was washed with sat. NaHCO3 soln. (3 Â 30 ml), dried (Na2SO4), and evaporated. The product was
purified by CC (AcOEt/CH2Cl2/Et3N 45 :45 :10). The resulting gum was dissolved in 2 ml of dry toluene, and the
soln. was added dropwise to stirred cold petroleum ether (100 ml). The solid product was filtered off: 7a,b.
2'-Deoxy-5'-O-(4,4'-dimethoxytrityl)-N6-(naphthalen-1-ylmethyl)adenosine 3'-(2-Cyanoethyl Diisopropyl-
phosphoramidite) (7a): 31P-NMR (CDCl3): 149.951 (94% pure). FAB-MS (CHCl3 ‡ 3-nitrobenzyl alcohol):
‡
894 ([M ‡ H] ).

Helvetica Chimica Acta ± Vol. 83(2000)
1415
2'-Deoxyy-5'-O-(4,4'-dimethoxytrityl)-N6-(pyren-1-ylmethyl)adenosine 3'-(2-Cyanoethyl Diisopropylphos-
phoramidite) (7b): 31P-NMR (CDCl3): 149.98 (99% pure). FAB-MS (CHCl3 ‡ 3-nitrobenzyl alcohol): 968
‡
([M ‡ H] ).
REFERENCES
[1] R. G. Harvey, ꢁPolycyclic Hydrocarbons and Carcinogenesisꢀ, American Chemical Society, Washington,
D.C., 1985.
[2] R. G. Harvey, N. E. Geacintov, Acc. Chem. Res. 1988, 21, 66.
[3] I. B. Weinstein, A. M. Jeffery, K. W. Jennette, S. H. Blobstein, R. G. Harvey, C. Harris, H. Autrup, H. Kasai,
K. Nakanishi, Science (Washington, D.C.) 1976, 193, 592; A. M. Jeffery, K. W. Jennette, S. H. Blobstein,
I. B. Weinstein, F. A. Beland, R. G. Harvey, H. Kasai, I. Miura, K. Nakanishi, J. Am. Chem. Soc. 1976, 98,
5714.
[4] A. M. Hruszkewycz, K. A. Canella, K. Peltonen, L. Kotrappa, A. Dipple, Carcinogenesis 1992, 13, 2347;
D. B. Reardon, A. C. H. Bigger, J. Strandberg, H. Yagi, D. M. Jerina, A. Dipple, Chem. Res. Toxicol. 1989,
2, 12.
[5] S. J. C. Wei, R. L. Chang, C.-Q. Wong, N. Bhachech, X. X. Cui, E. Hennig, H. Yagi, J. M. Sayer, D. M.
Jerina, B. D. Preston, A. H. Conney, Proc. Natl. Acad. Sci. U.S.A. 1991, 88, 11227; H. Rodriguez, E. L.
Loechler, Biochemistry 1993, 32, 1759; S. Shibutani, L. A. Margulis, N. E. Geacintov, A. P. Grollman,
Biochemistry 1993, 32, 7531.
[6] J. A. Ross, G. B. Nelson, K. H. Wilson, J. R. Rabinowitz, A. Galati, G. D. Stoner, S. Nesnow, M. Moss,
Cancer Res. 1995, 55, 1039; M. S. Greenblatt, W. P. Bennett, M. Hollstein, C. C. Harris, Cancer Res. 1994,
54, 4855.
[7] P. C. Zamecnik, M. L. Stephenson, Proc. Natl. Acad. Sci. U.S.A. 1978, 75, 280.
[8] S. T. Crooke, Biotechnology 1992, 10, 882.
 Á
Â
[9] C. Helene, J.-J. Toulme, Biochim. Biophys. Acta 1990, 1049, 99.
[10] E. Uhlmann, A. Peyman, Chem. Rev. 1990, 90, 543.
[11] H. M. Weintraub, Sci. Am. 1990, 262, 3 4.
[12] W. F. Lima, B. P. Monia, D. J. Ecker, S. M. Freier, Biochemistry 1992, 31, 12055.
[13] S. T. Cload, P. L. Richardson, Y.-H. Huang, A. Schepartz, J. Am. Chem. Soc. 1993, 115, 5005.
[14] D. J. Ecker, T. A. Vickers, T. W. Bruice, S. M. Freier, R. D. Jenison, M. Manoharan, M. Zounes, Science
(Washington, D.C.) 1992, 257, 958.
[15] T. M. Woolf, D. A. Melton, C. G. B. Jennings, Proc. Natl. Acad. Sci. USA 1992, 89, 7305.
 Á
[16] J.-C. François, N. T. Thuong, C. Helene, Nucleic Acids Res. 1994, 22, 3943.
[17] M. A. Rosen, D. J. Patel, Biochemistry 1993, 32, 6563.
[18] S. A. El-Kafrawy, M. A. Zahran, E. B. Pedersen, Acta Chem. Scand. 1999, 53, 280.
[19] A. R. Katritzky, X. Lan, J. Z. Yang, O. V. Denisko, Chem. Rev. 1998, 98, 409.
[20] R. G. Harvey, ꢁPolycyclic Aromatic Hydrocarbons: Chemistry and Cacinogenicityꢀ, Cambridge University
Press, Cambridge, England, 1991.
[21] S. C. Cheng, A. S. Prakash, M. A. Pigott, B. D. Hilton, J. Roman, H. Lee, R. G. Harvey, A. Dipple, Chem.
Res. Toxicol. 1988, 1, 216.
[22] H. Lee, E. Luna, M. Hinz, J. J. Stezowski, A. S. Kiselyov, R. G. Harvey, J. Org. Chem. 1995, 60, 5604.
[23] L. J. McBride, M. H. Caruthers, Tetrahedron Lett. 1983, 24, 245.
[24] H. Lee, M. Hinz, J. J. Stezowski, R. G. Harvey, Tetrahedron Lett. 1990, 31, 6773.
[25] R. P. Iyer, W. Egan, J. B. Regan, S. L. Beaucage, J. Am. Chem. Soc. 1990, 112, 1253.
[26] R. P. Iyer, L. R. Phillips, W. Egan, J. B. Regan, S. L. Beaucage, J. Org. Chem. 1990, 55, 4693.
[27] J. Jr. SantaLucia, Proc. Natl. Acad. Sci. U.S.A. 1998, 95, 1460.
[28] O. M. Ali, E. B. Pedersen, Acta Chem. Scand. 1999, 53, 497.
[29] M. L. Zapp, M. R. Green, Nature (London) 1989, 342, 714.
[30] T. J. Daly, K. S. Cook, G. S. Gray, T. E. Maione, J. R. Rusche, Nature (London) 1989, 342, 816.
[31] C. A. Rosen, E. Terwilliger, A. Dayton, J. G. Sodroski, W. A. Haseltine, Proc. Natl. Acad. Sci. U.S.A. 1988,
85, 2071.
[32] M. H. Malim, J. Hauber, S.-Y. Le, J. V. Maizel, B. R. Cullen, Nature (London) 1989, 338, 254.
[33] B. K. Felber, M. Hadzopoulou-Cladaras, C. Cladaras, T. Copeland, G. N. Pavlakis, Proc. Natl. Acad. Sci.
U.S.A. 1989, 86, 1495.
[34] M. Emerman, R. Vazeux, K. Peden, Cell 1989, 57, 1155.

1416
Helvetica Chimica Acta ± Vol. 83(2000)
[35] E. T. Dayton, D. M. Powell, A. I. Dayton, Science (Washington, D.C.) 1989, 246, 1625.
[36] M.-L. Hammarskjöld, J. Heimer, B. Hammarskjöld, I. Sangwan, L. Albert, D. Rekosh, J. Virol. 1989, 63,
1959.
[37] D. D. Chang, P. A. Sharp, Cell 1989, 59, 789.
[38] X. Lu, J. Heimer, D. Rekosh, M.-L. Hammarskjöld, Proc. Natl. Acad. Sci. U.S.A. 1990, 87, 7598.
[39] M. Hadzopoulou-Cladaras, B. K. Felber, C. Cladaras, A. Athanassopoulos, A. Tse, G. N. Pavlakis, J. Virol.
1989, 63, 1265.
[40] J. Kjems, M. Brown, D. D. Chang, P. A. Sharp, Proc. Natl. Acad. Sci. U.S.A. 1991, 88, 683.
[41] S. T. Cload, A. Schepartz, J. Am. Chem. Soc. 1994, 116, 43 7.
Received March 1, 2000