Nonnucleosidic base surrogates: The effect of 1,2-disubstituted phenanthrenes on DNA duplex stability

Article abstract of DOI:10.1002/hlca.200490250

Phenanthrene-1,2-dimethanol was incorporated into oligodeoxynucleotides via formation of phosphodiester bonds (cf. Scheme 1). If placed at internal positions in a DNA duplex, a strong reduction of duplex stability is observed (Table 1). Terminal attachment of stretches of phenanthrene residues, however, leads to a substantial increase in stability. The stabilization is attributed to a cooperative interaction of the phenanthrene residues of the two strands rather than to dangling end effects. Chimeric oligomers containing a stretch of six phenanthrene residues show two separate transitions (Table 2): one arising from the denaturation of the DNA stem (observable by a hyperchromic effect at 260 nm) and a second one from the denaturation of the phenanthrene part (observable by temperature-dependant gel mobility assays). Based on these findings, a model of the chimeric hybrids is proposed, in which the phenanthrene residues stack in a zipper-like manner on top of the DNA base pairs without disrupting the B-form of the DNA stem (see Fig. 7).

Full text of DOI:10.1002/hlca.200490250

2790
Helvetica Chimica Acta Vol. 87 (2004)
Nonnucleosidic Base Surrogates: The Effect of 1,2-Disubstituted
Phenanthrenes on DNA Duplex Stability
by Damian Ackermann and Robert H‰ner*
Department of Chemistry and Biochemistry, University of Berne, Freiestrasse 3, CH-3012 Bern
(phone: ‡41316314382; e-mail: robert.haener@ioc.unibe.ch)
Phenanthrene-1,2-dimethanol was incorporated into oligodeoxynucleotides via formation of phospho-
diester bonds (cf. Scheme 1). If placed at internal positions in a DNA duplex, a strong reduction of duplex
stability is observed (Table 1). Terminal attachment of stretches of phenanthrene residues, however, leads to a
substantial increase in stability. The stabilization is attributed to a cooperative interaction of the phenanthrene
residues of the two strands rather than to dangling end effects. Chimeric oligomers containing a stretch of six
phenanthrene residues show two separate transitions (Table 2): one arising from the denaturation of the DNA
stem (observable by a hyperchromic effect at 260 nm) and a second one from the denaturation of the
phenanthrene part (observable by temperature-dependant gel mobility assays). Based on these findings, a
model of the chimeric hybrids is proposed, in which the phenanthrene residues stack in a zipper-like manner on
top of the DNA base pairs without disrupting the B-form of the DNA stem (see Fig. 7).
1. Introduction. Modified oligonucleotides have found widespread applications as
diagnostic and research tools. Furthermore, the generation of defined molecular
structures by using nucleic acid like building blocks is a research topic of increasing
interest [1]. Among the many factors contributing to the formation and stabilization of
secondary and tertiary structures in nucleic acids, aromatic stacking interactions [2]
play a predominant role. The influence of nucleoside-linked aromatic and hetero-
aromatic base analogs on duplex stability and base pairing properties has been reported
[3 9] and reviewed [10]. Furthermore, dangling nucleotides [11 13] or other terminal
aromatic residues, such as quinolones [14], pyrene [15], and naphthalenediimides [15]
may substantially increase the stability of a DNA duplex. Furthermore, it has been
shown that the incorporation of aromatic building blocks via flexible, nonnucleosidic
linkers into DNA gives rise to stable duplexes in which the nonnatural building blocks
contribute to the duplex stability by aromatic stacking interactions [16 19]. Within the
context of developing nonnucleosidic DNA-like building blocks, we investigated the
1,2-disubstituted phenanthrene building block shown in Scheme 1. Model studies
suggested that this building block, which is amenable to oligomerization via
phosphodiester-bond formation, should fit well into the right-handed helical structure
of natural B-DNA. Here, we report the synthesis of the building block 1, its
incorporation into oligodeoxynucleotides, as well as its influence on the structure and
stability of DNA.
2. Results and Discussion. 2.1. Synthesis of the Phosphoramidite Building Block 1.
At 1308, 1-ethenylnaphthalene (2) was treated with 2 equiv. of dimethyl ethynedicar-
¹ 2004Verlag Helvetica Chimica Acta AG, Z¸rich

Helvetica Chimica Acta Vol. 87 (2004)
2791
Scheme 1. Illustration of Oligo(phenanthrylmethyl phosphates)
boxylate in o-xylene for 3 h [20] to give 3 in a moderate yield of 34% (Scheme 2).
Reduction of diester 3 with LiBH4 [21] followed by crystallization afforded the desired
diol 4 in 34% yield. Reaction with 4,4'-dimethoxytrityl chloride ((MeO)2TrCl) led to
the formation of a 4:1 mixture of the two regioisomers 5 and 6, which were separated
by column chromatography. The correct assignment of the regioisomeric structures was
achieved through nuclear Overhauser experiments (NOE, see Exper. Part) with the
chloroacetate 7, which was obtained by derivatization of 5. Treatment of 5 with 2-
cyanoethyl diisopropylphosphoramidochloridite gave the desired phosphoramidite 1.
Furthermore, 5 was immobilized on controlled-pore glass (CPG) according to the
literature [22]. The solid support 8 was obtained with a loading density of 26 mmol/g
(see Exper. Part).
2.2. Synthesis and Purification of DNA/Phenanthrene Hybrid Oligomers. The
phosphoramidite 1 was incorporated into a series of different chimeric oligodeox-
ynucleotides (summarized in Tables 1 and 2). Incorporation of the modified building
block proceeded without difficulties under the standard conditions used for automated
oligonucleotide synthesis. Oligomers were assembled on commercial CPG solid
supports except for 1b, 1c, 1f, 1g, 1h, 1k, and 1l, which were synthesized on the solid
support 8. After removal from the support and deprotection of the oligomers (10m
conc. aqueous ammonia, 558 for 16 h), the crude products where purified by HPLC.
Sequences containing no or a single phenanthrene building block were purified on
anion-exchange material. Purification of sequences containing two or more phenan-
threne residues was possible only on reversed-phase stationary phase (see Exper. Part).
After desalting, all compounds were characterized by ESI-MS and stored as stock
solutions in H2O.
2.3. Thermal Denaturation Experiments. The stability of the phenanthrene-modified
DNA hybrid duplexes was analyzed by temperature-dependent UV spectroscopy. The
samples were measured at 2 mm duplex concentration in a solution of 150 mm NaCl and
10 mm Tris ¥ HCl in H2O at pH 7.4 .Tm Values were taken from the maximum of the first
derivative of the melting curves. All UV melting curves of the duplexes listed in Table 1
had a sigmoidal shape and showed one single transition. Fig. 1 shows the denaturation

2792
Helvetica Chimica Acta Vol. 87 (2004)
Scheme 2. Synthesis of Phosphoramidite Building Block 1 and Solid Support 8
a) Dimethyl ethynedicarboxylate, o-xylene (ˆ1,2-dimethylbenzene), 1308, 3 h, recryst.; 34%. b)LibH4, MeOH,
THF, reflux, 1 h; 34%. c) (MeO)2Tr, Et3N, pyridine, 17 h, r.t.; 42% (5), 8% (6). d) 1. Chloroacetatic anhydride,
collidine (ˆ2,4,6-trimethylpyridine), -ch2Cl2, r.t., 15 min; 2. HCOOH/THF/EtOH 50 :35 :15, 20 min, r.t.; 32%.
i
e) CIP1Pr2N)OCH2CH2CN, Pr2NEt, CH2Cl2, r.t., 30 min; 88%. f) 1. [1,4-phenylenebis(oxy)bis[acetic acid],
i
BOP (ˆ(1H-benzotriazol-1-yloxy)tris(dimethylamino)phosphonium hexafluorophosphate), Pr2NEt, DMAP
(ˆ N,N-dimethylpyridin-A-amine), pyridine, r.t., 1 h; 2. LCAA-GPG iPr2NEt, BOP, MeCN, r.t., 20 h; 3. Ac2O,
pyridine, r.t., 2 h; loading: 26 mmol/g.
curves for Entries 1 5. Furthermore, heating and cooling curves were superimposable
for each duplex, which means that the melting and annealing processes are at
equilibrium under the experimental conditions.
The Tm data show that replacement of terminal base pairs with building block B
(Table 1, Entries 2 and 3) leads to an increase in the Tm (4.3 and 7.68 for one and three
pairs of B, resp.). In contrast, introduction of B residues in the middle (Entries 4 and 5)
results in a large decrease in duplex stability (À19.2 and À 33.58, resp.). Very similar
trends were observed for duplexes where only one modified strand was paired with the
complementary reference strand. One or three phenanthrene residues attached at the
end of one strand have a small stabilizing or destabilizing effect (Entries 6 8). On the
other hand, phenanthrene residues in the middle of the duplex (Entries 9 and 10) are
not tolerated and, hence, lead to a severe destabilization. Thus, the data summarized in

Helvetica Chimica Acta Vol. 87 (2004)
2793
Table 1. Melting Temperatures of Phenanthrene-Containing DNA Duplexes
Entry
1
Duplexa)
Tm [8]b)
DTm [8]
1a ¥ 2a
5' AGC TCG GTC ATC 3'
3' TCG AGC CAG TAG 5'
5' AGC TCG GTC ATB 1''
3' TCG AGC CAG TAB 2''
5' AGC TCG GTC BBB 1''
3' TCG AGC CAG BBB 2''
5' AGC TCG BTC ATC 3'
3' TCG AGC BAG TAG 5'
5' AGC TBB BTC ATC 3'
3' TCG ABB BAG TAG 5'
5' AGC TCG GTC ATC 3'
3' TCG AGC CAG TAB 2''
5' AGC TCG GTC ATB 1''
3' TCG AGC CAG TAG 5'
5' AGC TCG GTC ATC 3'
3' TCG AGC CAG BBB 2''
5' AGC TCG GTC ATC 3'
3' TCG AGC BAG TAG 5'
5' AGC TCG GTC ATC 3'
3' TCG ABB BAG TAG 5'
56.2
2
3
1b ¥ 2b
1c ¥ 2c
1d ¥ 2d
1e ¥ 2e
1a ¥ 2b
1b ¥ 2a
1a ¥ 2c
1a ¥ 2d
1a ¥ 2e
60.5
63.8
37.0
22.7
57.3
58.2
55.4
35.5
< 5.0
4.3
7.6
4
À 19.2
À 33.5
1.1
5
6
7
2.0
8
À 0.8
À 20.7
<À 51.2
9
10
a) B denotes the nonnucleosidic phenanthrene residue derived from phosphoramidite 1 (see Scheme 1); 1'' or 2''
denotes linkage to CH2O at C(1) or C(2), respectively, of the phenanthrene residue. b) Conditions for Tm
measurements: 2 mm duplex conc., 150 mm NaCl, 10 mm Tris ¥ HCl (pH 7.4); exper. error: Æ 0.58.
Table 1 allow the following conclusion: i) building block B leads to a strong decrease in
DNA duplex stability if incorporated in the middle of the duplex stem, and ii) terminal
attachment of B either pairwise or a single opposite to a natural nucleotide increases
duplex stability.
2.4. Effect of Terminal Phenanthrene Residues on Duplex Stability. Based on these
observations, we extended our studies to chimeric oligomers with terminally attached
phenanthrene units of type B (see Table 2). All melting curves show a single transition,
indicating a cooperative melting of the hybrids (Fig. 2). While the hybrids with one,
two, and three phenanthrene pairs show no difference between the heating and cooling

2794
Helvetica Chimica Acta Vol. 87 (2004)
Table 2. Melting Temperatures of DNAs Containing Terminal Phenanthrene Residues
Entry
11
Duplexa)
Tm [8]b)
DTm [8]
1r¥ 2r
5' AGC TCG 3'
20.3
3' TCG AGC 5'
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
1f ¥ 2f
1g ¥ 2g
1h ¥ 2h
1i ¥ 2i
5' AGC TCG B 1''
3' TCG AGC B 2''
5' AGC TCG BB 1''
3' TCG AGC BB 2''
5' AGC TCG BBB 1''
3' TCG AGC BBB 2''
5' AGC TCG BBB B 1''
3' TCG AGC BBB B 2''
5' AGC TCG BBB BB 1''
3' TCG AGC BBB BB 2''
5' AGC TCG BBB BBB 1''
3' TCG AGC BBB BBB 2''
5' AGC TCG B 1''
3' TCG AGC 5'
5' AGC TCG BB 1''
3' TCG AGC 5'
5' AGC TCG BBB 1''
3' TCG AGC 5'
5' AGC TCG BBB B 1''
3' TCG AGC 5'
5' AGC TCG BBB BB 1''
3' TCG AGC 5'
5' AGC TCG BBB BBB 1''
3' TCG AGC 5'
5' AGC TCG 3'
3' TCG AGC B 2''
5' AGC TCG 3'
3' TCG AGC BB 2''
5' AGC TCG 3'
3' TCG AGC BBB 2''
5' AGC TCG 3'
3' TCG AGC BBB B 2''
5' AGC TCG 3'
3' TCG AGC BBB BB 2''
5' AGC TCG 3'
27.9
7.6
36.9
16.6
34.3
14.0
36.2, 38.0c)
12.5, 30.4c)
29.6, 40.2c)
25.6
15.9, 17.7c)
À 7.8, 10.1c)
9.3, 19.9c)
5.3
1k ¥ 2k
1l ¥ 2l
1f ¥ 2r
1g ¥ 2r
1h ¥ 2r
1i ¥ 2r
1k ¥ 2r
1l ¥ 2r
1r¥ 2f
1r¥ 2g
1r¥ 2h
1r¥ 2i
1r¥ 2k
1r¥ 2l
24.3
4.0
23.8
3.5
n.o.
n.o.
n.o.
34.3
14.0
10.7
8.7
31.0
29.0
27.1
6.8
25.7
5.4
27.1
6.8
3' TCG AGC BBB BBB 2''
a) B denotes the nonnucleosidic phenanthrene residue derived from phosphoramidite 1 (see Scheme 1); 1''or 2''
denotes linkage to C(1) or C(2), respectively, of the phenanthrene residue. b) Conditions for Tm measurements:
2 mm duplex conc., 150 mm NaCl, 10 mm Tris ¥ HCl (pH 7.4); n.o. ˆ no transition observed. c) First value is taken
from cooling ramp, second value from heating ramp (for cases, in which hysteresis was observed).
cycles, a slight hysteretic effect (ca. 28) is observed with four phenanthrene pairs (Entry
15). This effect increases dramatically upon addition of further phenanthrene pairs.
Thus, large hysteretic effects were observed with hybrids containing five and six
phenanthrene pairs (Entries 16 and 17) even when small heating rates (0.28/min, data
not shown) were applied. The results of thermal denaturation studies are summarized
in Table 2. The Tm data show that the attachment of terminal phenanthrene pairs results
in a substantial increase in the melting temperature. Thus, the value gradually increases

Helvetica Chimica Acta Vol. 87 (2004)
2795
Fig. 1. UV Melting curves of the duplexes 1a ¥ 2a, 1b ¥ 2b, 1c ¥ 2c, 1d ¥ 2d, 1e ¥ 2e (Table 1, Entries 1 5). Conditions:
2 mm duplex concentration, 150 mm NaCl, and 10 mm Tris ¥ HCl in H2O (pH 7.4); heating and cooling rate: 0.58/
min.
from 20.38 to 38.08 upon addition of up to four phenanthrene pairs. The determination
of the Tm values for the hybrids with five and six phenanthrene pairs is complicated by
the relatively slow kinetics of the denaturation/renaturation process resulting in the
hysteretic effect mentioned above. In general, however, the trend indicates a leveling of
the Tm values.
The positive effect on duplex stability by dangling nucleotides [11] or aromatic
residues is well known [12][14]. It is possible that this sometimes very large effect is
responsible for the stabilization we observed with the phenanthrene-containing
duplexes. To further elucidate this question, we investigated the effect on duplex
stability of dangling phenanthrene units on either of the two strands. The denaturation
curves were measured under the same conditions as with the other oligomers. The
results are given in Table 2 (Entries 18 29). As expected, terminal attachment of
phenanthrene residues leads to an increase in duplex stability. The 5'-linked residues
(Entries 24 29) have a much more pronounced effect than the 3'-linked analogs
(Entries 18 23). The largest stabilization results from a single aromatic residue. Upon
addition of further dangling phenanthrene units, however, a steady reduction of Tm is
observed in both series. With more than three phenanthrene residues at the 3'-end, no
transition is observed anymore. The different trends are represented in Fig. 3 showing
the DTm values as a function of the number of attached phenanthrene units, either as
dangling residues or as phenanthrene pairs. The curves reveal that the stabilization
resulting from one phenanthrene pair (blue curve) is less than the sum of the
stabilizations by one 3'- and one 5'-dangling residues (red curve). Two, three, or four

2796
Helvetica Chimica Acta Vol. 87 (2004)
Fig. 2. UV Melting curves of the duplexes 1f ¥ 2f, 1g ¥ 2g, 1h ¥ 2h, 1i ¥ 2i, 1k ¥ 2k, 1l ¥ 2l (Table 2, Entries 12 17).
Conditions: 2 mm duplex concentration, 150 mm NaCl, and 10 mm Tris ¥ HCl (pH 7.4) in H2O; heating and
cooling rate: 0.58/min.
pairs, however, stabilize the duplex more than would be expected from the sum of the
individual contributions of the dangling residues. It can, thus, be concluded that the
stabilization by terminally attached phenanthrene pairs is the result of a cooperative
effect rather than of the sum of dangling-end contributions.

Helvetica Chimica Acta Vol. 87 (2004)
2797
Fig. 3. Plot of DTm values against the number n of attached phenanthrene residues. Curves show the effect of
terminal phenanthrene pairs (red), 3'-dangling phenanthrene residues (blue), 5'-dangling phenanthrene
residues (green), and the sum of 3'- and 5'-dangling phenanthrene residues (black).
2.5. Gel Mobility Assays. To obtain more information on the stability of DNAs with
terminal phenanthrene stretches, we investigated the migration behavior of the
corresponding duplexes under native conditions. Surprised by the large hysteretic
effect and the low Tm values of duplexes 1k ¥ 2k and 1l ¥ 2l we further investigated the
duplex stability by gel shift experiments. Fig. 4 shows the mobility on polyacrylamide-
gel electrophorese (PAGE) at room temperature of the six different hybrids 1f ¥ 2f,
1g ¥ 2g, 1h ¥ 2h, 1i ¥ 2i, 1k ¥ 2k, and 1l ¥ 2l in comparison with the nonmodified reference
duplex 1r¥ 2r (see Table 2, Entries 11 17). After staining with ethidium bromide, all
lanes exhibited one band, indicating the oligomers are migrating as pairs. The obvious
gradual decrease in migration velocity, caused by each additional pair of phenanthrene
residues, is due to the increase in molecular mass and thus well in agreement with the
expectations.
The duplex 1l ¥ 2l, composed of six nucleotide and six phenanthrene residues
(Table 2, Entry 17) was further analyzed by temperature-dependent gel mobility assays.
PAGEs were run at three different temperatures, 25, 50, and 708 under nondenaturing
conditions (Fig. 5). At 25 and 508, the two complementary strands 1l ¥ 2l clearly migrate
as one single band, corresponding to a duplex. At 708, however, two bands
corresponding to the two single strands are visible. Thus, the duplex 1l ¥ 2l is stable
up to a temperature of 508. This does not correlate with the Tm values of 29.6 or 40.28
obtained from the UV transition curves (see Table 2, Entry 17 and Fig. 2). Obviously,
the transition observed in the UV melting curve does not arise from the melting of the
whole duplex. Rather, that observed transition corresponds to the melting of the DNA

2798
Helvetica Chimica Acta Vol. 87 (2004)
Fig. 4. PAGEof phenanthrene-modified DNA duplexes . The number of phenanthrene pairs attached to the
DNA stem gradually increases from 0 to 6 (from left to right). Bands were visualized by staining with ethidium
bromide.
Fig. 5. PAGEanalysis of duplex 1l ¥ 2l and the single strands 1l and 2l at different temperatures. Bands were
visualized under UV light.
part of the chimeric duplex. A second transition, i.e., the denaturation of the
phenanthrene part, occurs at higher temperature but is not observable by measurement
of UV absorption, probably due to a lack of hyperchromicity.

Helvetica Chimica Acta Vol. 87 (2004)
2799
2.6. CD Spectroscopy of the Phenanthrene-Modified Duplexes. Fig. 6 shows the
circular dichroism (CD) spectra of the duplexes 1a ¥ 2a, 1b ¥ 2b, 1c ¥ 2c, 1d ¥ 2d, 1e ¥ 2e
(Table 1, Entries 1 5) measured at 58. All other conditions were identical to the
thermal denaturation experiments. The spectra of all hybrids show a minimum CD
effect in the range of 250 Æ 10 nm and a maximum in the range of 275 Æ 10 nm, which is
very characteristic for B-form DNA. An intact B-form is, thus, maintained within the
DNA parts of the different hybrids. In addition, a maximum CD effect is observed at
240 245 nm for the duplex 1c ¥ 2c and a very strong one for duplex 1e ¥ 2e. Both of
these hybrids contain three phenanthrene units. This maximum is not observed for
duplexes having only a single phenanthrene pair, and it is much more pronounced when
the phenanthrene stretch is embedded within the DNA core rather than attached at the
end of the DNA stem. This suggests that stretches of phenanthrene residues although
achiral building blocks by themselves contribute to the secondary structure of the
hybrid formed by the chimeric oligomers.
Fig. 6. CD Spectra of the duplexes 1a ¥ 2a, 1b ¥ 2b, 1c ¥ 2c, 1d ¥ 2d, and 1e ¥ 2e. Conditions: 2 mm duplex
concentration, 150 mm NaCl, and 10 mm Tris ¥ HCl (pH 7.4) in H2O; 58.
2.7. Model of a DNA Duplex Containing Terminal Phenanthrene Pairs. A model of
the 1g ¥ 2g duplex, composed of six DNA base pairs and two phenanthrene pairs is
shown in Fig. 7. The conformation represents a local-minimum structure obtained with
Hyperchem starting from B-form DNA by using the AMBER2 force-field. The
following findings were taken into account: i) the DNA stem maintains an intact B-
form; ii) phenanthrene units of the two strands contribute to hybrid stability through
cooperative p-stacking interactions between the strands, and iii) despite the achiral
nature of the building blocks, phenanthrene stretches contribute to the secondary

2800
Helvetica Chimica Acta Vol. 87 (2004)
structure of the hybrid formed by the chimeric oligomers. Thus, the model illustrates
the interstrand stacking of the phenanthrene residues on top of a B-form DNA. The top
view (right part of Fig. 7) shows the arrangement of the phenanthrene residues along
the helical axis. The phenanthrene residues are stacking in a zipper-like fashion on top
of the bases of the DNA stem without disrupting its B-form structure.
Fig. 7. AMBER2-Minimized model of duplex 1g ¥ 2g. The four phenanthrene-1,2-dimethyl moieties are
highlighted in non-CPK-colors. Left: view perpendicular to the helical axis; right: view along the helical axis.
3. Conclusions. Phenanthrene-1,2-dimethanol was synthesized and evaluated as a
nonnucleosidic, non-H-bonding DNA-base surrogate. The building block strongly
disfavors duplex formation when placed at internal positions of a DNA stem. The
destabilization is likely due to a bad geometric match of the two different stacking
systems (i.e., DNA and phenanthrene stretches). Terminal attachment of phenan-
threne residues, however, leads to a substantial increase in stability. The stabilization is
due to a cooperative interaction of the phenanthrene units of the two strands rather
than to dangling-end effects. The DNA part of the chimeric hybrids maintains the
common B-form structure. Chimeric oligomers containing a stretch of six phenan-
threne residues show two separate transitions: one arising from the denaturation of the
DNA stem (observable by a hyperchromic effect at 260 nm) and a second one from the
denaturation of the phenanthrene part (observable by temperature-dependent gel
mobility assays). Based on these findings, a model of the chimeric hybrids is proposed,
in which the phenanthrene residues stack in a zipper-like way on top of the DNA base
pairs without disrupting the B-form of the DNA stem.

Helvetica Chimica Acta Vol. 87 (2004)
2801
Experimental Part
1. General. The 2-cyanoethyl diisopropylphosphoramidochloridite was prepared as described in [23]. If not
indicated otherwise, chemicals and solvents for reactions were purchased from Fluka, Acros, or Aldrich and
were used without further purification. TLC: silica gel SIL G-25 UV254 glass plates (Macherey-Nagel); UV
detection and/or dipping in a soln. of anisaldehyde/H2SO4/EtOH 1:1:18, followed by heating. Flash column
chromatography (FC): silica gel 60 (63 32 mm, Chemie Brunschwig AG). CD Spectra: Jasco J-715
spectropolarimeter with a 150-W Xe high-pressure lamp and a Jasco PDF-350S-Peltier unit, coupled with a
Colora K5 ultrathermostat, for the control of the temp. of the cell holder; temp. determination directly in the
sample. 1H- and 13C-NMR: Bruker AC-300; d values in ppm (solvent signals as internal references), J in Hz. 31P-
NMR: Bruker AMX-400; d values in ppm (85% H3PO4 as external reference). ESI-MS: VG-Platform single-
quadrupole ESI spectrometer; in m/z.
2. Phosphoramidite Building Block 1 and Solid Support 8. 1-Ethenylnaphthalene (2). After stirring a
t
suspension of BuOK (6.3 g, 56 mmol) and MePPh3Br (20 g, 56 mmol) in THF (200 ml) for 10 min at r.t.,
naphthalene-1-carboxaldehyde (6.0 g, 37 mmol) was added. The mixture was stirred for 60 min at r.t., then the
reaction was quenched by adding 10% citric acid soln. in H2O (200 ml) and, after evaporation of THF, the
residue was extracted with hexane (3Â). The org. phase was dried (MgSO4) and evaporated and the residue
adsorbed on SiO2 (30 g). Purification by FC (SiO2 20 g), hexane ! hexane/AcOEt 9 :1) gave 2 (5.5 g, 95%).
Colorless oil. TLC (hexane/AcOEt 9 :1): Rf 0.90.
Phenanthrene-1,2-dicarboxylic Acid Dimethyl Ester (3). A soln. of 2 (5.5 g, 35.7 mmol) and dimethyl
ethynedicarboxylate (17.7 g, 125 mmol) in o-xylene (80 ml) was stirred for 3 h at 1308. All volatile compounds
were evaporated, and the residue was adsorbed on SiO2 (20 g). Purification by FC (SiO2 30 g), hexane !
hexane/AcOEt 6 :4) gave 3 (3.4g, 34%). Colorless solid. TLC (hexane/AcOEt 8 :2): Rf 0.35. 1H-NMR
(300 MHz, CDCl3): 3.99, 4.11 (2s, 2 Me); 7.68 7.74( m, HÀC(3), HÀC(6), HÀC(7)); 7.85 (d, J ˆ 9.0, HÀC(9));
7.93 (dd, J ˆ 2.5, 8.0, HÀC(8)); 8.23 (d, J ˆ 9.0, HÀC(10)); 8.71 (dd, J ˆ 2.1, 8.0, HÀC(5)); 8.80 (d, J ˆ 8.7,
HÀC(4)). 13C-NMR (75 MHz, CDCl3): 52.7, 53.0 (2q, Me); 121.31, 121.35, 124.0 (3d, arom. C); 125.1 (s, arom.
C); 126.3, 127.4, 128.17 (3d, arom. C); 128.21 (s, arom. C); 128.2, 128.7, 129.1 (3d, arom. C); 129.4, 132.5, 133.3,
‡
134.9 (4s, arom. C); 166.1, 169.8 (2s, C(O)). ESI-MS (pos.): 317.0783 (C18H14O4Na ; calc. 317.0789).
Phenanthrene-1,2-dimethanol (4). A soln. of 3 (2.2 g, 7.5 mmol) and MeOH (900 ml, 22 mmol) in THF
(30 ml) was carefully treated with LiBH4 (490 mg, 22 mmol) at 48. After refluxing for 1 h, the mixture was
cooled in an ice bath, and the reaction was quenched by the addition of H2O (5 ml) followed by careful addition
of 1n HCl (50 ml). The mixture was extracted with CH2Cl2 (3 Â 50 ml) and the org. phase dried (MgSO4) and
evaporated. The residue was suspended in AcOEt (20 ml) and the suspension refluxed for 30 min, diluted with
hexane (10 ml), and cooled to r.t. The fine white crystals were filtered and washed with AcOEt/hexane 1 :1: 4
(768 mg, 43%). TLC (hexane/AcOEt 1:1): Rf 0.15. 1H-NMR (300 MHz, CDCl3): 4.88, 5.12 (2s, CH2OH); 7.47
7.59 (m, HÀC(6), HÀC(7)); 7.66 (d, J ˆ 8.7, HÀC(3)); 7.75 (d, J ˆ 9.2, 1 arom. H); 7.81 (dd, J ˆ 1.5, 8.5,
HÀC(8)); 8.16 (d, J ˆ 9.4, 1 arom. H); 8.66 (d, J ˆ 8.1, HÀC(5)); 8.68 (d, J ˆ 8.7, HÀC(4)). 13C-NMR (75 MHz,
CDCl3): 61.0, 66.1 (2d, CH2OH); 127.5, 128.2, 128.8, 131.74, 131.78, 131.8, 131.9, 133.4 (8d, arom. C); 134.3, 135.2,
‡
135.8, 136.1, 140.0, 144.3 (6s, arom. C). ESI-MS (pos.): 261.0884(C 16H4O2Na ; calc. 261.0891).
2-{[(4,4'-Dimethoxytrityl)oxy]methyl}phenanthrene-1-methanol (5) and 1-{[(4,4'-dimethoxytrityl)oxy]-
methyl}phenanthrene-2-methanol (6). A soln. of 4 (610 mg, 2.5 mmol) in pyridine (10 ml) was treated with
Et3N (1.40 ml, 10 mmol) and (MeO)2TrCl (1.21 g, 3.6 mmol), stirred overnight at r.t., and then diluted with
CH2Cl2 (50 ml) and 10% citric acid soln. in H2O (50 ml). The mixture was extracted with CH2Cl2 (3Â), the org.
phase dried (MgSO4) and evaporated, and the residue adsorbed on SiO2 (4g). Purification by two FCs SiO 2
(40 g); hexane ! hexane/AcOEt 6 :4; then SiO2 (20 g); hexane/AcOEt 9 :1 ! hexane/AcOEt 8 :2) gave 5
(585 mg, 42%) and 6 (110 mg, 8%).
Data of 5: TLC (hexane/AcOEt 7:3): Rf 0.50. 1H-NMR (300 MHz, CDCl3): 2.75 (br. s, OH); 3.80 (s,
MeO); 4.42 (s, ArCH2C); 4.95 (s, CH2OH); 6.85 6.91 (m, 4arom. H); 7.23 7.36 ( m, 3 arom. H); 7.43 7.46 (m,
4arom. H); 7.51 7.69 ( m, 5 arom. H); 7.84( d, J ˆ 9.3, HÀC(9)); 7.91 (dd, J ˆ 1.4, 7.6 HÀC(8)); 8.22 (d, J ˆ 9.3,
HÀC(10)); 8.69, 8.71 (2d, J ˆ 8.5, 8.1, HÀC(4), HÀC(5)). 13C-NMR (75 MHz, CDCl3): 55.2 (q, Me); 58.4, 65.9
(2d, CH2); 87.6 (s, Ar3CO); 113.4, 122.6, 122.8, 123.0, 126.72, 126.75, 127.0, 127.8, 127.94, 128.01, 128.15, 128.5,
130.0 (13d, arom. C); 130.4, 130.7, 131.1, 131.6, 135.5, 135.8, 136.6, 144.7, 158.6 (9s, arom. C). ESI-MS (pos.):
‡
563.2192 (C37H32O4Na ; calc. 563.2198).
Data of 6: TLC (hexane/AcOEt 1:1): Rf 0.25. 1H-NMR (300 MHz, CDCl3): 2.74(br. s, OH), 3.71 (s,
MeO); 4.58, 4.61 (2s, ArCH2O); 6.77 6.82 (m, 4arom. H); 7.15 7.27 ( m, 4arom. H); 7.37 7.64( m, 9 arom. H);
7.74 7.82 ( m, 2 arom. H); 8.60 8.65 (m, HÀC(4), HÀC(5)). 13C-NMR (75 MHz, CDCl3): 55.3 (q, MeO); 59.3,

2802
Helvetica Chimica Acta Vol. 87 (2004)
63.9 (2d, CH2); 87.3 (s, Ar3CO); 113.4, 122.9, 123.1, 123.4, 126.7, 127.0, 127.2, 128.0, 128.2, 128.3, 128.5, 130.2,
130.41 (13d, arom. C); 130.45, 131.1, 131.4, 132.6, 135.7, 139.5, 144.6, 158.7 (8s, arom. C). ESI-MS (pos.):
‡
563.2213 (C37H32O4Na ; calc. 563.2198).
Chloroacetic Acid [2-(Hydroxymethyl)phenanthren-1-yl]methyl Ester (7).
A soln. of 5 (100 mg,
0.186 mmol), collidine (40 mg, 0.33 mmol), and chloroacetic anhydride (38 mg, 0.22 mmol) in CH2Cl2
(0.4ml) was stirred for 15 min at r.t. The reaction was quenched by the addition of 10% citric acid soln. in
H2O (50 ml), the mixture extracted with CH2Cl2 (3 Â 50 ml), and the org. phase dried (MgSO4) and
evaporated. The crude was dissolved in HCOOH/THF/EtOH 50 :35 :15 (v/v; 6 ml), and the soln. was stirred for
20 min at r.t. After addition of 1m NH4Cl in H2O (10 ml), THF and EtOH were evaporated. The residue was
extracted with CH2Cl2 (3Â), the org. phase dried (MgSO4) and evaporated, and the crude adsorbed on SiO2
(300 mg). Purification by FC (SiO2 (2 g); hexane ! hexane/AcOEt 8 :2) gave 7 (18 mg, 32%). White solid. TLC
(hexane/AcOEt 7:3): Rf 0.15. 1H-NMR (400 MHz, CDCl3): 4.06 (s, CH2Cl); 5.05 (d, J ˆ 5.6, CH2OH); 5.92 (s,
ArCH2O); 7.62 7.71 (m, HÀC(6), HÀC(7)); 7.74( d, J ˆ 8.6, HÀC(3)); 7.86, 8.04(2 d, J ˆ 9.3, HÀC(9),
HÀC(10)); 7.92 (dd, J ˆ 1.1, 7.4, HÀC(8)); 8.70 (d, J ˆ 8.2, HÀC(5)); 8.77 (d, J ˆ 6.6, HÀC(4)). NOE (irr.
Signal ! affected signals): 5.05 (CH2OH) ! 2.32 (d, OH; 1.6%), 5.92 (d, ArCH2OC; 3.0%), and 7.74( d,
HÀC(3); 3.7%); 5.92 (ArCH2OC) ! 5.05 (d, CH2OH; 4.5%) and 8.04 (d, HÀC(10); 8.4%); 8.04 (HÀC(10)) !
5.92 (d, ArCH2OC; 8.7%), 7.86 (d, HÀC(9); 2.4%), and 7.92 (neg. d, artefact), but no effect at 8.77 (d, J ˆ 6.6,
HÀC(4)). 13C-NMR (75 MHz, (D6)MSO): 46.4, 65.8, 66.1 (3d, CH2); 127.7, 128.2, 129.3, 131.8, 132.1, 132.3,
132.9 (7d, arom. C); 133.4( s, arom. C); 133.5 (d, arom. C); 134.4, 135.0, 136.0, 136.1, 146.1 (5s, arom. C); 172.6 (s,
‡
CˆO). ESI-MS (pos.): 314.0710 (C18H15ClO3 ; calc. 314.0710).
{2-{[(4,4'-Dimethoxytrityl)oxy]methyl}phenanthren-1-yl}methyl Diisopropylphosphoramidite (1). A soln.
of 5 (340 mg, 0.6 mmol) and iPr2NEt (270 ml, 1.6 mmol) in CH2Cl2 (4ml) was treated with 2-cyanoethyl
diisopropylphosphoramidochloridite (200 mg, 0.8 mmol), stirred for 2 h at r.t., and then purified by FC SiO2
(6 g), hexane (‡2% Et3N) ! hexane/AcOEt 7:3 (‡2% Et3N): 1 (390 mg, 88%). White foam. TLC (hexane/
1
AcOEt 7:3): Rf 0.65. H-NMR (300 MHz, CDCl3): 0.92, 1.05 (2d, J ˆ 6.8, (Me2CH)2N); 2.35 (dt, J ˆ 3.2, 1.1,
OCH2CH2CN); 3.47 3.59 (m, (Me2CH)2N, OCH2CH2CN); 3.73 (s, MeO); 4.51 (s, ArCH2OC); 4.97 5.15 (m,
ArCH2OP); 6.84 6.89 ( m, 4arom. H); 7.21 7.36 ( m, 3 arom. H); 7.42 7.47 (m, 4arom. H); 7.53 7.69 ( m, 4
arom. H); 7.80 (d, J ˆ 9.2, HÀC(9)); 7.90 (d, J ˆ 8.5, HÀC(8)); 8.15 (d, J ˆ 9.2, HÀC(10)); 8.72, 8.75 (2d, J ˆ
9.0, 8.8, HÀC(4), HÀC(5)). 13C-NMR (75 MHz, CDCl3): 20.1 (t, J(C,P) ˆ 6.8, CH2CN); 24.50, 24.60, (2q,
J(C,P) ˆ 7, ( Me2CH)2N); 43.2 (d, J(C,P) ˆ 12, (Me2CH)2N); 55.2 (q, MeO); 58.4, 58.8 (2d, J(C,P) ˆ 20, 17,
ArCH2OP, POCH2CH2CN); 64.0 (d, ArCH2OC); 86.8 (s, Ar3CO); 113.2 (d, arom. C); 117.6 (s, CN); 122.7, 123.1,
123.31, 123.34, 126.5, 126.6, 126.7, 126.8, 127.2, 127.9, 128.3, 128.4, 130.2 (13d, arom. C); 130.4, 131.1, 131.5, 132.5,
132.6, 136.2, 136.3, 145.1, 158.5 (9s, arom. C). 31P-NMR (121 MHz, CDCl3): 149.2. ESI-MS (pos.): 763.3259
‡
(C46H49N2O5PNa ; calc. 763.3276).
Solid support 8 (ˆ{4-{2-[(Controlled-pore-glass)amino]-2-oxoethoxy}phenoxy}acetic Acid {2-{[(4,4'-Di-
methoxytrityl)oxy]methyl}penanthren-1-yl}methyl Ester). A soln. of 5 (116 mg, 0.21 mmol), [1,4-phenylenebi-
s(oxy)]bis[acetic acid] (97 mg, 0.43 mmol), DMAP (5 mg, 0.04 mmol), and iPr2NEt (185 ml, 1.1 mmol) in
pyridine (2 ml) was treated with BOP (190 mg, 0.43 mmol) and kept for 30 min at r.t. The mixture was diluted
with CH2Cl2 (20 ml) and 10% citric acid soln. in H2O (30 ml). After extraction with CH2Cl2 (3 Â 20 ml), the
org. phase was washed with sat. aq. NaHCO3 soln., dried (MgSO4) and evaporated and the crude purified by FC
‡
(SiO2 (1 g), CH2Cl2 (‡2% Et3N) ! CH2Cl2/MeOH 19 :1 (‡2% Et3N)) to give the intermediate as Et3NH salt.
1H-NMR (300 MHz, CDCl3): 1.25 (t, J ˆ 7.4, ( Me2CH2)3N); 2.99 (q, J ˆ 7.4, (Me2CH2)3N); 3.69 (s, MeO); 4.32,
4.34, 4.35, 5.54 (4s, CH2); 6.62 6.79 (m, 8 arom. H); 7.13 7.78 (m, 15 arom. H); 8.61 8.69 (m, HÀC(4),
HÀC(5)).
The intermediate (25 mg, 0.03 mmol) was dissolved in a suspension of (long-chain-alkyl)amino controlled-
i
pore glass (LCAA-CGP; 500 mg), Pr2NEt (60 ml, 0.36 mmol), and BOP (13 mg, 0.03 mmol) in MeCN and
shaken for 20 h at r.t. After filtration, the solid phase was washed with MeCN and CH2Cl2, suspended in pyridine
(1.5 ml) and Ac2O (0.5 ml), and shaken for 2 h at r.t. The solid was filtered and washed with DMF and CH2Cl2.
Loading density (trityl monitoring): 26 mmol/g.
3. Synthesis and Purification of Oligonucleotides. Oligonucleotides were synthesized on a 392-DNA/RNA
synthesizer (Applied Biosystems) according to the phosphoramidite chemistry [24][25]. The deoxynucleoside
phosphoramidites were from ChemGenes (Ashland, MA). Oligonucleotides were prepared by the standard
synthetic procedure (−trityl-off× mode), by using 1.1m tert-butyl hydroperoxide (in hexane/1,2-dichlorethane 1:4)
as the oxidant. For the phenanthrenylmethyl phosphoramidite 1, used as a 0.1m soln. in 1,2-dichloroethane, the
coupling time was increased to 50 s in the presence of 0.25m 5-(benzylthio)-1H-tetrazole in MeCN as activator.
Cleavage from the solid support and final deprotection was achieved by treatment with 30% NH4OH soln.

Helvetica Chimica Acta Vol. 87 (2004)
2803
overnight at 558. The nonmodified oligonucleotides were purified anion by exchange HPLC, (Source 15Q 4.6/
100 PE (Pharmacia); eluent A, 20 mm Na2HPO4 (pH 11.5); eluent B, 20 mm NaH2PO4, 2m NaCl (pH 11.5)).
The modified oligonucleotides were purified by reversed-phase HPLC (LiChrospher 100 RP-18 (5 mm; Merck);
eluent A, 0.1m (Et3NH)OAc in H2O (pH 7.0); eluent B, MeCN; elution at 508). The purified oligonucleotides
were desalted with Sep-Pak cartridges (Waters, Milford, USA; eluent A, 0.1m (Et3NH)OAc; eluent B, H2O/
MeCN 1:1). The molecular masses of the purified oligonucleotides were determined by ESI-MS. The data of all
oligomers are summarized In Table 3:
Table 3. ESI-MS and Other Data of the Synthesized Oligomers
Sequencea)
A.u. [OD]
e [cmÀ1 mÀ1
]
MS [m/z]b)
calc.
found
1a
2a
1b
2b
1c
2c
1d
2d
1e
2e
1r
2r
1f
2f
1g
2g
1h
2h
1i
5' AGC TCG GTC ATC 3'
3' TCG AGC CAG TAG 5'
5' AGC TCG GTC ATB 1∫
3' TCG AGC CAG TAB 2∫
5' AGC TCG GTC BBB 1∫
3' TCG AGC CAG BBB 2∫
5' AGC TCG BTC ATC 3'
3' TCG AGC BAG TAG 5'
5' AGC TBB BTC ATC 3'
3' TCG ABB BAG TAG 5'
5' AGC TCG 3'
3' TCG AGC 5'
5' AGC TCG B 1''
3' TCG AGC B 2''
5' AGC TCG BB 1''
3' TCG AGC BB 2''
5' AGC TCG BBB 1''
3' TCG AGC BBB 2''
15
27
66
20
55
22
40
24168500
38
31
29
31
99
27
51
40
122300
132900
157900
164200
219600
225900
153600
3681
3621
3670
3632
3641
3616
36243624
3592
3620
3670
3631
3640
3615
3591
3681
220500
235400
62500
3575
3664
1792
1792
2093
2093
2393
2393
2693
2693
2993
2993
3293
3574
3662
1793
1792
2093
2093
2393
2393
2694
2693
2993
2993
3294
62500
105500
105500
148500
148500
191500
191500
234500
234500
277500
3293
55
27
67
36
5' AGC TCG BBB B 1''
3' TCG AGC BBB B 2''
5' AGC TCG BBB BB 1''
3' TCG AGC BBB BB 2''
5' AGC TCG BBB BBB 1''
3' TCG AGC BBB BBB 2''
2i
1k
2k
1l
25
34277500
35
31
3293
320500
320500
3593
3593
3594
3594
2l
a) B is the nonnucleosidic phenanthrene unit; 1'' or 2'' denotes linkage to CH2O at C(1) or C(2), respectively of
b
the phenanthrene unit. ) Calculated for [M À H]À.
4. Thermal Denaturation Experiments. All experiments were carried out under the following conditions:
2.0 mm oligonucleotide concentration (each strand), 10 mm Tris ¥ HCl buffer (pH 7.4), and 150 mm NaCl. UV
Melting Curves: Varian Cary-3e-UV/VIS spectrophotometer equipped with a Peltier-block temperature
controller. The Varian WinUV software were utilized to determine the melting curves at 260 nm, a heating-
cooling-heating cycle in the temp. range of 0 808 was applied with a temp. gradient of 0.58/min. To avoid H2O
condensation on the UV cells at << 208, the cell compartment was flushed with N2. Data were collected and
¾
analyzed with Kaleidagraph software from ¹Synergy Software. Tm Values were determined as the maximum of
the first derivative of the melting curve.
5. PAGE Experiments. Equipment. Mini-Protean III (Bio-Rad); EPS-3500 XL (Pharmacia Biotech) power
supply. Buffer: 5  TBE (ˆ450 mm Tris 450 mm boric acid 10 mm EDTA).
Preparation of 16% Gels. An aq. soln. of 40% acrylamide/bisacrylamide 19 :1 (4 ml) was diluted with 5 Â
TBE (2 ml) and H2O (4ml). After degassing the acrylamide soln., a soln. of 10% APS (ammonium persulfate)
in H2O (20 ml) and TMEDA (tetramethylethylenediamine; 20 ml) was added, and the gel was allowed to
polymerize within 1 h.

2804
Helvetica Chimica Acta Vol. 87 (2004)
Running the Gels. Loading of oligonucleotide (0.1 OD) in formamide/H2O 1 :1 (6 ml) per well. Elution
buffer: 1 Â TBE. Electrophoresis was carried out at 150 200 V for 20 min at 25, 50 or 708. The bands of the gels
were directly photographed under UV light or stained with ethidium bromide.
6. Modelling. The structure of the duplex 1g ¥ 2g (see Fig. 7) was minimized by using the AMBER2 force
field (Hyperchem 7.0, Hypercube, Waterloo, Ontario).
REFERENCES
[1] J. Wengel, Org. Biomol. Chem. 2004, 2, 277.
[2] C. A. Hunter, K. R. Lawson, J. Perkins, C. J. Urch, J. Chem. Soc., Perkin Trans. 2 2001, 651.
[3] I. Singh, W. Hecker, A. K. Prasad, S. P. A. Virinder, O. Seitz, Chem. Commun. 2002, 500.
[4] M. Berger, A. K. Ogawa, D. L. McMinn, Y. Q. Wu, P. G. Schultz, F. E. Romesberg, Angew. Chem., Int. Ed.
2000, 39, 2940.
[5] D. L. McMinn, A. K. Ogawa, Y. Q. Wu, J. Q. Liu, P. G. Schultz, F. E. Romesberg, J. Am. Chem. Soc. 1999,
121, 11585.
[6] R. X. F. Ren, N. C. Chaudhuri, P. L. Paris, S. Rumney, E. T. Kool, J. Am. Chem. Soc. 1996, 118, 7671.
[7] J. M. Gao, C. Strassler, D. Tahmassebi, E. T. Kool, J. Am. Chem. Soc. 2002, 124, 11590.
[8] C. Brotschi, A. Haberli, C. J. Leumann, Angew. Chem., Int. Ed. 2001, 40, 3012.
[9] G. Mathis, J. Hunziker, Angew. Chem., Int. Ed. 2002, 41, 3203.
[10] E. T. Kool, J. C. Morales, K. M. Guckian, Angew. Chem., Int. Ed. 2000, 39, 990.
[11] S. Bommarito, N. Peyret, J. SantaLucia, Nucl. Acids. Res. 2000, 28, 1929.
[12] K. M. Guckian, B. A. Schweitzer, R. X. F. Ren, C. J. Sheils, P. L. Paris, D. C. Tahmassebi, E. T. Kool, J. Am.
Chem. Soc. 1996, 118, 8182.
[13] A. A. Mokhir, C. Richert, Nucleic Acids Res. 2000, 28, 4254.
[14] J. Tuma, W. H. Connors, D. H. Stitelman, C. Richert, J. Am. Chem. Soc. 2002, 124, 4236.
[15] J. Michel, K. Bathany, J. M. Schmitter, J. P. Monti, S. Moreau, Tetrahedron 2002, 58, 7975.
[16] S. M. Langenegger, R. H‰ner, Helv. Chim. Acta 2002, 85, 3414.
[17] S. M. Langenegger, R. H‰ner, Chem. Biodiv. 2004, 1, 259.
[18] C. B. Nielsen, M. Petersen, E. B. Pedersen, P. E. Hansen, U. B. Christensen, Bioconjug. Chem. 2004, 15,
260.
[19] U. B. Christensen, M. Wamberg, F. A. El Essawy, A. Ismail, C. B. Nielsen, V. V. Filichev, C. H. Jessen, M.
Petersen, E. B. Pedersen, Nucleosides Nucleotides Nucleic Acids 2004, 23, 207.
[20] C. M. Chan, Synth. Commun. 1989, 19, 1981.
[21] K. Soai, A. Ookawa, J. Org. Chem. 1986, 51, 4000.
[22] R. T. Pon, S. Y. Yu, Nucleic Acids Res. 1997, 25, 3629 3635.
[23] M. V. Rao, C. B. Reese, V. Schehlmann, P. S. Yu, J. Chem. Soc., Perkin Trans.1 1993, 4 3.
[24] S. L. Beaucage, M. H. Caruthers, Tetrahedron Lett. 1981, 22, 1859.
[25] N. D. Sinha, J. Biernat, J. McManus, H. Koster, Nucleic Acids Res. 1984, 12, 4539.
Received July 16, 2004