Intercalating nucleic acids containing insertions of naphthalimide

Article abstract of DOI:10.1002/hlca.200690177

In a study of linker-length dependence, we evaluated naphthalimide (= 1H-benzo[de]isoquinoline-1,3(2H)-dione) and 4-bromonaphthalimide as intercalating nucleic acids. We used a vicinal dihydroxy system when incorporating the six different naphthalimide mo

Full text of DOI:10.1002/hlca.200690177

1826
Helvetica Chimica Acta – Vol. 89 (2006)
Intercalating Nucleic Acids Containing Insertions of Naphthalimide
by Michael C. Wamberg, Krzysztof Walczak1), Lars Andersen, Allam A. Hassan2), and Erik B. Pedersen*
Nucleic Acid Center3), Department of Physics and Chemistry, University of Southern Denmark,
Campusvej 55, DK-5230 Odense M
(phone: +4565502555; fax: +4566158780; e-mail: ebp@chem.sdu.dk)
In a study of linker-length dependence, we evaluated naphthalimide (=1H-benzo[de]isoquinoline-
1,3(2H)-dione) and 4-bromonaphthalimide as intercalating nucleic acids. We used a vicinal dihydroxy
system when incorporating the six different naphthalimide monomers into DNA, and found the mini-
mum linker-length to be five C-atoms. With this length of the linker, naphthalimide was discriminating
between DNA and RNA – stabilizing DNA, while destabilizing RNA. Furthermore, naphthalimide
showed universal base character by hybridizing to the four natural bases with a range as narrow as
1.48. When compared to pyrene, naphthalimide with the same linker-length gave significantly higher ther-
mal meltings when hybridized to DNA.
1. Introduction.
belongs to a group of promising DNA-targeting anticancer agents [1]. Bisnaphthal-
ACHTREUNGimides Elinafide (1) [2] and Bisnafide (2) [3][4] consisting of two naphthalimide resi-
– Naphthalimide (=1H-benz[de]isoquinoline-1,3(2H)-dione)
dues connected by an aminoalkyl linker have been reported to inhibit topoisomerase
II [5] by acting as bisintercalators in the major groove of DNA [6][7]. These symmet-
rical compounds are currently undergoing clinical trials [8–10].
The most active mononaphthalimides Amonafide (3) and Mitonafide (4) [1][11]
have also been reported to inhibit topoisomerase II [12][13]. They were selected for
clinical trials. However, the development of Mitonafide was stopped in phase II due
to lack in efficacy in the solid tumors that were tested [14], while Amonafide completed
clinical phase I and II trials [1]. Amonafide (3) and Mitonafide (4) differ only in the sub-
stituent at C(5), having an NH2 and a NO2 group, respectively.
The potency as antitumor agents made it interesting for us to test naphthalimide as
an intercalating nucleic acid, which we define as an oligonucleotide with a covalently
bound intercalator. Besides the type of aromatic compound used as intercalator, impor-
tant factors to consider, when using intercalating nucleic acids for hybridization, are the
structure of the backbone and the length of the linker. Ikeda et al. [15] used PNA [16]
when incorporating naphthalimide as an intercalator into DNA. We chose a vicinal
dihydroxy system for bulge incorporations, which we have previously described for pyr-
1
)
On leave from: Department of Organic Chemistry, Biochemistry and Biotechnology, Silesian Uni-
versity of Technology, Krzywoustego 4, PL-44-100 Gliwice.
Current address: Faculty of Education, Suez Canal University, Egypt.
A research center funded by The Danish National Research Foundation for studies on nucleic acid
chemical biology.
2
3
)
)
ꢀ 2006 Verlag Helvetica Chimica Acta AG, Zürich

Helvetica Chimica Acta – Vol. 89 (2006)
1827
ene as the intercalator [17]. Intercalating nucleic acids containing bulge insertions of
(R)-1-O-[(pyren-1-yl)methyl]glycerol (INAꢁ; 5) showed discrimination between
DNA and RNA [17], as significantly by increased affinities for ssDNA were achieved,
while INA/RNA duplexes were destabilized. The fluorescence properties of the pyrene
intercalator confirmed its intercalation rather than groove-binding in duplexes [18].
These findings led to commercialization of INAꢁ [19].
In this paper, we present the synthesis of six intercalating nucleic acids with naph-
thalimide or 6-bromonaphthalimide derivatives 6–11 as the intercalator and their eval-
uation by thermal stability measurements on DNA/DNA and DNA/RNA duplexes.
For monomers 6–9, we have varied the configuration of the stereogenic center in
the linker and the substitution at C(6) of the naphthalimide ring, in order to evaluate
the dependence on configuration for the linker. Filichev et al. [20] found (R)-1-O-
[(pyren-1-yl)methyl]glycerol to give higher stabilization than its corresponding (S)-iso-
mer. Monomers 10 and 11 were synthesized only as (S)-isomers, because that would

1828
Helvetica Chimica Acta – Vol. 89 (2006)
give the same spatial configuration as the (R)-isomer of 1-O-[(pyren-1-yl)methyl]gly-
cerol.
2. Results and Discussion. – 2.1. Chemistry. Both 4-bromonaphthalene-1,8-dicar-
boxylic anhydride (12) and naphthalene-1,8-dicarbocylic anhydride (13) were reacted
with the (R)- and (S)-form of 3-aminopropane-1,2-diol (14a and 14b, resp.) to give
the monomers 6–9 in 60–89% yield (Scheme 1). The diols were prepared for oligonu-
cleotide synthesis using standard protecting groups. The primary alcohol was protected
by using 4,4’-dimethoxytrityl chloride in dry pyridine to give 5’-O-DMT-protected com-
pounds 15a–15d in 67–75% yield. The secondary alcohol was phosphitylated with (2-
cyanoethoxy)bis(diisopropylamino)phosphane in dry CH2Cl2 to yield 16a–16d in
22–98%.
Scheme 1
For monomers 10 and 11 another synthetic route was used (Scheme 2). Enantiomer-
ically pure (S)-1,2-O-isopropylidene-4-O-(methylsulfonyl)butane-1,2,4-triol (17) [21]
was obtained from (S)-malic acid in four steps. (S)-Malic acid (=hydroxybutanedioic
acid) was converted to dimethyl (S)-malate according to Mori and Ikunaka [22]. The
diester was reduced with LiAlH4 to (S)-butane-1,2,4-triol and protected with an isopro-
pylidene group as described by Hayashi et al. [23]. Mesylation of (S)-1,2-O-isopropy
idenebutane-1,2,4-triol was carried out either in pyridine according to Augustyns et al.
[21] or in CH2Cl2/Et3AN according to Kim et al. [24]. Using pyridine as the solvent, it
gave 17 in a yield of 51%, while the latter solvent gave 17 in 99% yield.
ACHTRElUNG -
CTHREUNG

Helvetica Chimica Acta – Vol. 89 (2006)
1829
Scheme 2
Enantiomerically pure (S)-1,2-O-isopropylidene-5-O-(methylsulfonyl)pentane-
1,2,5-triol (18) [25] was obtained from L-glutamic acid in four steps. L-Glutamic acid
was converted to (S)-2,3,4,5-tetrahydro-5-oxofuran-2-carboxylic acid as described by
Herdeis [26]. Reduction with LiAlH4 according to Brunner and Lautenschlager [27]
yielded (S)-pentane-1,2,5-triol, which was protected with an isopropylidene group
and mesylated as described above to give 18.
For N-alkylation, naphthalimide (19) was deprotonated with NaH in DMF, and, in
the presence of the phase-transfer catalyst Bu
18 to give 2-{2-[(S)-2,2-dimethyl-1,3-dioxolan-4-yl]ethyl}-1H-benz[de]
NBr (TBAB), reacted with either 17 or
isoquinoline-
AHCTREUNG
1,3(2H)-dione (20) and 2-{3-[(S)-2,2-dimethyl-1,3-dioxolan-4-yl]propyl}-1H-benz[de]-
isoquinoline-1,3(2H)-dione (21) in 53–54% yield. Treatment of 20 and 21 with 80%
aqueous AcOH gave the diols 10 and 11, respectively. These were DMT-protected
using DMT-Cl in pyridine or in CH2Cl2/Et3ACHTERNUNG , and phosphitylated as described above
to yielding 24 and 25, respectively.
Prior to oligonucleotide synthesis, we studied the stability of the 6-Br substituent on
naphthalimide under the alkaline conditions normally used for the deprotection of the
protecting groups on the nucleobases (NH3, 558, overnight). No NH2 substitution was
observed (monitored by TLC and NMR) under these conditions.
The DMT-protected phosphoramidites of the intercalating nucleic acid monomers
6–11 were incorporated into DNA oligonucleotides using the same coupling times (2
min) in the oligonucleotide synthesis as was used for the amidites of natural nucleo-

1830
Helvetica Chimica Acta – Vol. 89 (2006)
sides. The monomers were inserted as a bulge in a dodecamer highly conserved HIV-1
long-term repeat region [28], which we have previously used for pyrene-intercalating
nucleic acids [17][18]. All modified oligonucleotides were confirmed by MALDI-
TOF analysis (Table 1).
Table 1. Masses of ODNs
Monomer 5’-CTCAAGXCAAGCT-3’ 5’-CTCAAXGXCAAGCT-3’ 5’-CTCAXAGXCAAGCT-3’ 5’-CTCAXAGCAXAGCT-3’
Calc.
Found
Calc.
Found
Calc.
Found
Calc.
Found
6
7
8
9
10
11
4025
4025
3946
3946
3960
3974
4026
4027
3946
3945
3960
3973
4436
4436
4278
4278
4306
4334
4438
4436
4279
4279
4309
4334
4436
4436
4278
4278
–
4440
4438
4280
4281
–
4436
4436
4278
4278
–
4440
4438
4282
4282
–
4334
4334
4334
4333
2.2. Thermal Melting Studies. The project was designed to study the linker-length
dependence for intercalating nucleic acids with naphthalimide as the intercalating moi-
ety. It is clearly seen from Table 2 (Entry 1) that all intercalators with short linkers, i.e.,
6–10, are destabilizing the DNA/DNA duplex. Only 11, having a linker of five C-
atoms, was able to stabilize the duplex. Studies of the configuration in the linker
showed that the (R)-isomers, i.e., 6 and 8, were better than their corresponding (S)-iso-
mers 7 and 9. Introduction of a Br substituent at C(6) of naphthalimide did not have an
effect on the (R)-isomer, while DTm for the (S)-isomer was increased by 1.48 (7 com-
pared to 9).
Table 2. Melting Temperatures of DNA/DNA Duplexes with X=6–11 inserted as a Bulge. The target
sequence was 5’-AGC TTG CTT GAG-3’, and the reference melting temperature for the wild-type
duplex without insertions was 47.48. Data of X=5 [17] are given for comparison.
Tm [8]
X=
3’-TCG AAC XGA ACT C-5’
3’-TCG AAC XGX AAC TC-5’
3’-TCG AAC XGA XAC TC-5’
3’-TCG AXA CGA XAC TC-5’
5 [17]
6
7
8
9
10
11
50.4
51.4
55.6
60.8
45.4
44.1
46.7
50.9
42.4
34.1
40.0
42.9
45.4
40.4
44.1
49.6
41.0
32.5
31.9
32.4
46.1
44.9
–
53.0
55.5
62.7
62.4
–
From Table 3, it is seen that all naphthalimides – independent of the length of the
linker – destabilized the DNA/RNA duplex. Surprisingly, 10 was the most destabilizing
intercalator for the DNA/RNA duplex, when only one insertion was made. It was
expected that an intercalator with a linker of four C-atoms would be better than one
with a three C-atom linker. (R)-Isomers, i.e., 6 and 8, were in general slightly favored
over (S)-isomers 7 and 9, and, in general, the 6-Br-substituted naphthalimides had
higher Tm values than the unsubstituted ones. Intercalating nucleic acids containing
insertions of 11 were discriminating between DNA and RNA, as the DNA/DNA
duplexes were stabilized while DNA/RNA duplexes were destabilized.

Helvetica Chimica Acta – Vol. 89 (2006)
1831
Table 3. Melting Temperatures of DNA/RNA Duplexes with X=6–11 Inserted as a Bulge. The target
sequence was 5’-AGC UUG CUU GAG-3’, and the reference melting temperature for the wild-type
duplex without insertions was 40.58. Data of X=5 [17] are given for comparison (reference for X=5 was
42.28 [17]).
Tm [8]
X=
3’-TCG AAC XGA ACT C-5’
3’-TCG AAC XGX AAC TC-5’
3’-TCG AAC XGA XAC TC-5’
3’-TCG AXA CGA XAC TC-5’
5 [17]
6
7
8
9
10
11
37.8
34.2
32.6
35.0
35.4
28.5
30.7
34.8
34.9
27.8
34.4
34.3
33.0
28.2
29.2
35.0
31.8
27.9
26.7
28.8
31.7
30.7
–
38.0
38.4
35.8
36.2
–
Gallego and Reid [7] have shown that the bisintercalator 1 is stacking mainly with G
and A bases. So, to further investigate the discrimination and stabilization phenomena,
intercalating nucleic acids with two naphthalimides were prepared (Table 2, Entries
2–4). The monomers were inserted with varying distances as next-nearest neighbors
or with two or four nucleobases between them. Placed between two adenines four
nucleobases apart, the naphthalimide (R)-isomers 6 and 8 with the shorter linker
were actually able to stabilize the DNA/DNA duplex. In all other cases with short link-
ers of (R)-isomers, destabilization was observed when compared with the wild-type
duplex. Also insertion of the (S)-isomers did not lead to any stabilization. In fact, dou-
ble insertions of 9 led a further destabilization of ca. 108 when compared with a single
insertion. It was interesting to observe 11 giving stabilizations up to 15.38 (7.78 per mod-
ification), when two insertions were made.
Compared to previous studies [17] of intercalating nucleic acids with pyrene as the
intercalator (data are given for comparison in Tables 2 and 3), 11 gives higher thermal
melting values than pyrene 5.
For pyrene 5, better stacking was achieved when two insertions were made between
two adenines four nucleobases apart, but 11 was equally good, whether incorporated
between the adenines four bases apart or incorporated two nucleobases apart with
one insertion between two adenines and the other between a CG base pair (Table 2,
Entries 3–4).
2.3. Universal Base Studies. We found it interesting to test naphthalimide as a uni-
versal base. Results are presented in Table 4. Position 7 in the 13-mer duplex was varied
with all four natural bases in both strands to give a reference data set. Then INAꢁ (5)
and ODNs with naphthalimide intercalators 6–11 inserted at position 7 were measured
against oligonucleotides with all four natural bases.
From Table 4, it is obvious that the Br-substituted naphthalimides 6 and 7 were good
universal bases having a narrow range of melting temperatures against all natural bases,
whereas those with the same linker-length but without the Br substituents, i.e., 8 and 9,
showed a broad range of melting temperatures. However, 10 and 11 having the longer
linkers were the best ones. For 11, the range was as narrow as 1.48, and, furthermore, the
melting temperatures were higher than for the bases with the shorter linkers, and they
were only by 58 lower than for the wild-type duplex. For INAꢁ (5), the range was 2.68.
2.4. Mismatch Studies. Finally, a series of mismatch studies was carried out (Table 5).
The specificity for hybridization was measured by the difference in the melting temper-

1832
Helvetica Chimica Acta – Vol. 89 (2006)
Table 4. Universal Base Measurements. Melting temperatures of duplexes where a natural nucleotide was
replaced by either another nucleoside or by monomers 5–11.
5’-AGCTTGYCTTGAG-3’
3’-TCGAACXGAACTC-5’
X=
A
C
G
T
5
6
7
8
9
10
11
Y
A
C
G
T
39.4
35.5
45.9
48.9
37.4
23.6
52.4
32.5
45.8
54.3
46.7
36.1
50.5
35.1
43.1
35.8
43.1
44.1
41.5
43.1
38.9
36.7
36.9
39.6
40.4
38.1
38.7
41.7
36.8
33.8
32.0
36.8
37.9
34.3
34.1
39.2
40.8
37.9
38.3
39.4
42.1
42.6
41.2
42.0
Table 5. Melting Temperatures [8] of Mismatched sequences with 6–11 Inserted as a Bulgea)
Sequence: 5’-AGC TTZ YTT GAG-3’
3’-TCG AAG X GAA CTC-5’
X=
Z Y Tm DTm Tm DTm Tm DTm Tm DTm Tm DTm Tm DTm Tm DTm
Wild type G C 47.4 45.4 42.1 45.4 41.0 46.1 53.0
–
6
7
8
9
10
11
Mut. 1
Mut. 2
Mut. 3
Mut. 4
Mut. 5
Mut. 6
C C 23.6 ꢀ23.8 34.1 ꢀ11.3 32.5 ꢀ9.6 30.2 ꢀ15.2 32.2 ꢀ8.8 32.7 ꢀ13.4 32.7 ꢀ20.3
A C 30.7 ꢀ16.7 32.2 ꢀ13.2 31.5 ꢀ10.6 29.8 ꢀ15.6 31.2 ꢀ9.8 32.9 ꢀ13.2 35.3 ꢀ17.7
T C 28.9 ꢀ18.5 32.0 ꢀ13.4 31.3 ꢀ10.8 30.0 ꢀ15.4 31.2 ꢀ9.8 32.5 ꢀ13.6 36.7 ꢀ16.3
G G 39.8 ꢀ7.6 36.3 ꢀ9.1 32.5 ꢀ9.6 33.3 ꢀ12.1 32.5 ꢀ8.5 33.5 ꢀ12.6 39.7 ꢀ13.3
G A 39.0 ꢀ8.4 34.7 ꢀ10.7 32.8 ꢀ9.3 32.2 ꢀ13.2 33.2 ꢀ7.8 34.8 ꢀ11.3 38.6 ꢀ14.4
G T 35.9 ꢀ11.5 37.5 ꢀ7.9 36.4 ꢀ5.7 34.0 ꢀ11.4 33.0 ꢀ8.0 36.5 ꢀ9.6 43.1 ꢀ9.9
a) DTm is the difference in Tm between the matched sequence and the mismatched one.
ature between the fully complementary duplex and the duplex where one mismatch has
been introduced in position either Y or Z.
It is interesting to observe 9 showing almost equal specificity for all mismatches.
Unfortunately, the specificity was rather low, being in the range of 7.8–9.88, lower
than the matched sequence. Compound 11 was the most sensitive monomer to mis-
matches especially at the 3’-site of the intercalator, where it was in the range of the
unmodified sequence. Therefore, we decided to expand the studies, performing mis-
match measurements with double insertions of 11 as next-nearest neighbors, and
with two and four base pairs in-between, respectively (Table 6). Again, one mismatch
was introduced in position either Y or Z.
Results for probes II and III with the intercalators separated by two and four base
pairs, respectively, gave almost identical results. These probes showed in many cases
even higher specificity. The CꢀC mismatch causes a drop in melting temperature as
high as ca. 298.
2.5. Molecular Modelling. To gain further insight into the linker-length dependence
of intercalating nucleic acids, molecular modelling studies were performed with Macro-
Model version 5.1.016 from Schrçdinger Inc. [29]. All calculations were conducted with
OPLS-AA force field [30] and the GB/SA water model [31]. From the Figure,b, it is

Helvetica Chimica Acta – Vol. 89 (2006)
1833
Table 6. Melting-Temperature [8] Data of Mismatched Sequences with Bulge Insertions of 11 as Next-
Nearest Neighbors, and with Two and Four Base Pairs in-between, Respectivelya)
Target: 5’-AGC TTZ YTT GAG-3’
Probes:
I 3’-TCGAAC X GXAA CTC-5’
II 3’-TCG AAC X GAXA CTC-5’
III 3’-TCG AXAC GAXA CTC-5’
X=
I
II
III
Z Y
Tm
DTm
Tm
DTm
Tm
DTm
Wild type
Mut. 1
Mut. 2
Mut. 3
Mut. 4
Mut. 5
Mut. 6
G C
C C
A C
T C
G G
G A
G T
55.5
38.8
38.9
38.6
38.8
37.5
44.9
62.7
33.1
41.3
38.8
52.0
50.7
46.1
62.4
33.4
40.4
37.9
52.2
50.7
45.6
ꢀ16.7
ꢀ16.6
ꢀ16.9
ꢀ16.7
ꢀ18.0
ꢀ10.6
ꢀ29.6
ꢀ21.4
ꢀ23.9
ꢀ10.7
ꢀ12.0
ꢀ16.6
ꢀ29.0
ꢀ22.0
ꢀ24.5
ꢀ10.2
ꢀ11.7
ꢀ16.8
a) DTm is the difference in Tm between the matched sequence and the mismatched one.
clearly seen that 8 having the shortest linker is unable to position the intercalator opti-
mally for base stacking with nucleobases. Furthermore, the backbone of the strand con-
taining the intercalator is disturbed by the intercalator pulling it towards the axe of the
duplex. Comparing the linker-length of three and four C-atoms (Fig.,b and c), it is seen
that the four C-atom linker positions the intercalator deeper into the duplex. The back-
bone conformation looks similar to that of DNA without intercalator (Fig.,a). Compar-
ing the linker-length of four and five C-atoms (Fig.,c and d) reveals that the intercalator
stays in the same position between the nucleobases, but the backbone is pushed a little
backwards. Thus, the linker of four C-atoms is sufficient to position the intercalator
between the nucleobases and achieve stacking. By increasing the length of the linker
to five C-atoms, a better stacking is not achieved. However, more favorable thermal
meltings were obtained with the five C-atom linker, and this may reflect that the back-
bone becomes more relaxed with a longer linker.
3. Conclusions. – We have tested naphthalimide as intercalating nucleic acid in a
study of linker-length dependence. This dependence was obvious as shorter linkers
destabilized the duplexes since they were not able to position the intercalator optimally
for base stacking without disturbing the backbone, which was confirmed by molecular-
modelling studies. The study revealed that naphthalimide must have a linker of at least
five C-atoms, i.e., 11. Furthermore, we found this naphthalimide 11 to have a higher
thermal melting than INAꢁ (5) when hybridized to the complementary DNA strand.
Naphthalimide is discriminating between DNA and RNA like INAꢁ (5), as DNA is
stabilized while RNA is destabilized. Naphthalimide shows universal-base character,
hybridizing to all four natural bases within a range as narrow as 1.48. Finally, 11 showed
sensibility to mismatches especially if two naphthalimides were incorporated with two
or four base pairs in-between.

1834
Helvetica Chimica Acta – Vol. 89 (2006)
Figure. Molecular modelling conformations of a DNA/DNA duplex (a) with insertions of one
naphthalimide intercalator (b–d) with a linker of three (b), four (c), and five (d) C-atoms, respectively.
AHCTREUNG

Helvetica Chimica Acta – Vol. 89 (2006)
1835
Dr. Vyacheslav Filichev is acknowledged for the preparation of INAꢁ used for universal-base stud-
ies.
Experimental Part
General. TLC: TLC plates 60 F254 (Merck); visualized by UV light (254 nm) or a stain of
(NH4)6Mo7O24 ·4 H2O/Ce2(SO4)3 (50 :1 (v/v)) in 5% H2SO4. Column chromatography (CC): silica-gel-
packed column (silica gel 60; 0.040–0.063 mm, Merck). Solvents used for CC were distilled prior to
use, while reagents were used as purchased. Petroleum ether: b.p. 60–808. M.p.: Büchi melting-point
apparatus. NMR Spectra: Varian Gemini 2000 spectrometer (1H: 300 MHz, 13C: 75 MHz, 31P: 121.5
1
MHz); d values in ppm relative to Me
Si as internal standard for H-NMR; CDCl3 (d 77.00), DMSO
(d 39.52), and CD3A
CN (d 1.32) for 13C-NMR; 85% H3PO4 as external standard for 31P-NMR, J in Hz.
EI-MS: Finnigan Mat SSQ 701 mass spectrometer. MALDI-MS: Accurate ion mass determinations
were performed using an Ionspec 4.7 T Ultima Fourier Transform mass spectrometer (IonSpec, Irvine,
CA). The M+ or [M+Na]+ ions were peak-matched using ions derived from the 2,5-dihydroxybenzoic
acid matrix. Elemental analyses were performed at H. C. Ørsted Institute, University of Copenhagen.
General Procedure for the Synthesis of 6-Bromo-2-(2,3-dihydroxypropyl)-1H-benzo[de]isoquinoline-
1,3(2H)-dione (6 and 7) and 2-(2,3-dihydroxypropyl)-1H-benzo[de]isoquinoline-1,3(2H)-dione (8 and 9).
To a suspension of 4-bromonaphthalene-1,8-dicarboxylic anhydride (12; 1.16 g, 4 mmol) or naphthalene-
1,8-dicarboxylic anhydride (13; 0.80 g, 4 mmol) in anh. EtOH (60 ml) was added 3-aminopropane-1,2-
diole (14a or 14b; 0.36 g, 4 mmol), and the reaction mixture was refluxed. After 30 min, the suspension
was dissolved completely, and, after 2.5 h, TLC (5% MeOH/CHCl3) indicated complete consumption of
the substrate. The warm soln. was filtered off, and, after cooling, precipitation occurred. The product was
recrystallized from EtOH.
6-Bromo-2-[(R)-2,3-dihydroxypropyl]-1H-benzo[de]isoquinoline-1,3(2H)-dione (6). Yield: 1.24 g
(89%). M.p. 179–1808 (EtOH). [a]D =+39.4 (c=0.7, MeOH). 1H-NMR (DMSO): 3.34–3.44 (m, 2
HꢀC(3)); 3.87–3.98 (m, HbꢀC(1), HꢀC(2)); 4.18 (dd, J=7.5, 12.5, HaꢀC(1)); 4.60 (t, J=5.6, HOꢀ
C(3)); 4.77 (d, J=5.0, HOꢀC(2)); 7.91 (dd, J=7.3, 8.4, 1 arom. H); 8.10 (d, J=8.1, 1 arom. H); 8.22
(d, J=8.1, 1 arom. H); 8.39 (dd, J=0.8, 8.4, 1 arom. H); 8.46 (d, J=7.3, 1 arom. H). 13C-NMR
(DMSO): 43.36 (C(1)); 64.56 (C(3)); 68.47 (C(2)); 121.92, 122.70, 128.04, 128.57, 128.83, 129.50,
130.71, 131.14, 131.34, 132.23 (arom. C); 162.93, 162.99 (2 C=O). EI-MS: 350 (6, M+), 352 [M+2]+,
333 (24), 320 (93), 318 (100), 289 (40), 276 (38), 260 (25), 126 (60). Anal. calc. for C15H12BrO4N
(350.17): C 51.45, H 3.45, N 4.00; found: C 51.45, H 3.39, N 3.98.
6-Bromo-2-[(S)-2,3-dihydroxypropyl]-1H-benzo[de]isoquinoline-1,3(2H)-dione (7). Yield: 1.23 g
(88%). M.p. 179–1808. [a]D =ꢀ40.6 (c=0.7, MeOH). Spectroscopic data identical to those of 6.
2-[(R)-2,3-Dihydroxypropyl)-1H-benzo[de]isoquinoline-1,3(2H)-dione (8). Yield: 0.77 g (72%). Rf
1
(5% MeOH/CHCl3) 0.16. M.p. 145–1498. H-NMR (DMSO): 3.44 (m, 2 HꢀC(3)); 3.94–4.03 (m, Hbꢀ
C(1), HꢀC(2)); 4.23 (dd, J=8.0, 12.2, HaꢀC(1)); 4.61, (br. s, OH), 4.80 (br. s, OH), 7.86 (m, 2 arom.
H), 8.36–8.56 (m, 4 arom. H). 13C-NMR (DMSO): 43.20 (C(1)); 64.60 (C(3)); 68.56 (C(2)); 122.18,
127.09, 127.31, 127.49, 130.53, 131.17, 131.29, 132.42, 134.03, 135.34 (arom. C); 163.63 (C=O). HR-
MALDI-MS: 294.0726 ([M+Na]+, C NaNO4þ ; calc. 294.0736).
19ACHTERNUGH13ACHTREGUN
2-[(S)-2,3-Dihydroxypropyl)-1H-benzo[de]isoquinoline-1,3(2H)-dione (9). Yield: 0.64 g (60%). Rf
1
(5% MeOH/CHCl3) 0.20. M.p. 147–1518. H-NMR (CDCl3): 3.20, 3.37 (2 br. s, 2 OH); 3.67 (m, 2 Hꢀ
C(3)); 4.11 (m, HꢀC(2)); 4.34 (dd, J=6.4, 15.7, HaꢀC(1)); 4.44 (dd, J=5.0, 15.7, HbꢀC(1)); 7.72 (t,
J=8.0, 2 arom. H); 8.16 (d, J=8.0, 2 arom. H); 8.55 (d, J=7.4, 2 arom. H). 13C-NMR (CDCl3): 42.61
(C(1)); 63.82 (C(3)); 70.60 (C(2)); 121.93, 126.96, 127.38, 127.99, 131.39, 131.68, 133.30, 134.39, 135.25
(arom. C); 165.24 (C=O). HR-MALDI-MS: 294.0730 ([M+Na]+, C NaNO4þ ; calc. 294.0736).
15ACHTERNUGH13ACHTREGUN
Anal. calc. for C15H13NO4 (294.07): C 66.41, H 4.83, N 5.16; found: C 66.52, H 4.54, N 4.71.
General Procedure for DMT-Protection of Diols 6–9. Diols 6–9 was dissolved in anh. pyridine
(10–25 ml) and 4,4’-dimethoxytrityl chloride (1.5 equiv.) was added while stirring. After 24 h, the reac-
tion was stopped by addition of MeOH (2 ml) and the solvent was removed by evaporation under

1836
Helvetica Chimica Acta – Vol. 89 (2006)
reduced pressure. The residue was dissolved in CH2Cl2 (50 ml), washed with sat. aq. NaHCO3 (2”5 ml),
dried (Na2SO4) and evaporated to dryness. The product was purified by CC (silica gel; cyclohexane/
AcOEt/Et3ACHTRENUGN (2 :8 :0.5 v/v/v)).
2-{(R)-3-[Bis(4-methoxyphenyl)(phenyl)methoxy]-2-hydroxypropyl}-6-bromo-1H-benzo[de]isoqui-
noline-1,3(2H)-dione (15a). Yield: 0.49 g (75%). Colorless solid (obtained from 0.35 g, 1.0 mmol of 6).
1H-NMR (CDCl3): 2.87 (d, J=7.1, OH); 3.27 (dd, J=4.5, 9.5, HbꢀC(3)); 3.34 (dd, J=4.5, 9.5, Haꢀ
C(3)); 3.77 (s, 2 MeO); 4.17–4.18 (m, HꢀC(2)); 4.27 (dd, J=9.0, 13.7, HbꢀC(1)); 4.59 (dd, J=9.0,
13.7, HaꢀC(1)); 6.81 (d, J=8.4, 4 arom. H); 7.21–7.48 (m, 9 arom. H); 7.82 (dd, J=7.1, 8.5, 1 arom.
H); 8.01 (d, J=8.1, 1 arom. H); 8.38 (d, J=8.1, 1 arom. H); 8.54 (dd, J=1.2, 8.5, 1 arom. H); 8.62 (dd,
J=1.1, 7.4, 1 arom. H). 13C-NMR (CDCl3): 43.98 (C(1)); 55.15 (MeO); 65.64 (C(2)); 69.91 (C(3));
86.09 (Ar3C); 113.06, 122.00, 122.86, 126.72, 127.78, 128.05, 128.14, 128.99, 130.04, 130.44, 130.55,
131.07, 131.39, 132.22, 133.39, 135.89, 135.91, 144.68, 158.40 (arom. C); 164.36 (C=O).
2-{(S)-3-[Bis(4-methoxyphenyl)(phenyl)methoxy]-2-hydroxypropyl}-6-bromo-1H-benzo[de]isoqui-
noline-1,3(2H)-dione (15b). Yield: 0.45 g (67%). Colorless solid (obtained from 0.35 g, 1.0 mmol of 7).
Spectroscopic data identical to those of 15a.
2-{(R)-3-[Bis(4-methoxyphenyl)(phenyl)methoxy]-2-hydroxypropyl}-1H-benzo[de]isoquinoline-1,3-
ACHTREUNG
CTHREUNG
HꢀC(1)); 3.77 (s, 2 MeO); 4.18 (m, HꢀC(2)); 4.31 (dd, J=13.6, 3.2, HbꢀC(3)); 4.59 (dd, J=13.6, 9.2,
HaꢀC(3)); 6.80 (dd, J=8.9, 2.7, 4 arom. H); 7.18 (t, J=7.2, 1 arom. H); 7.26 (t, J=7.2, 2 arom. H);
7.36 (dd, J=8.9, 1.5, 4 arom. H); 7.48 (d, J=7.2, 2 arom. H); 7.72 (dd, J=7.3, 8.2, 2 arom. H); 8.19
(dd, J=8.2, 0.9, 2 arom. H); 8.57 (dd, J=7.3, 0.9, 2 arom. H). 13C-NMR (CDCl3): 43.94 (C(1)); 55.15
(MeO); 65.70 (C(3)); 70.10 (C(2)); 86.03 (Ar3C); 113.05, 122.41, 126.68, 126.88, 126.99, 127.40, 127.76,
128.14, 129.08, 130.03, 131.39, 131.49, 133.33, 134.05, 135.25, 135.94, 135.97, 144.74, 158.37 (arom. C);
164.96 (C=O). HR-MALDI-MS: 596.2038 ([M+Na]+, C NaNO6þ ; calc. 596.2043). Anal. calc. for
36ACHTERNUGH31ACHTREGUN
C36H31NO6 (596.20): C 75.38, H 5.45, N 2.44; found: C 75.87, H 5.40, N 2.05.
2-{(S)-3-[Bis(4-methoxyphenyl)(phenyl)methoxy]-2-hydroxypropyl}-1H-benzo[de]isoquinoline-1,3-
ACHTREUNG
CTHREUNG
CHRTEUNGH CHTREUNG
C 75.38, H 5.45, N 2.44; found: C 74.71, H 5.36, N 1.98.
General Procedure for Synthesis of Phosphoramidites 16a–16d: Compound 15a–15d (1 equiv.) was
dissolved in anh. CH2Cl2 (10 ml). (2-Cyanoethoxy)bis(diisopropylamino)phosphane (1 equiv.) was
added, followed by addition of N,N-diisopropylammonium tetrazolide (1.5 equiv.). The mixture was stir-
red at r.t. for 12 h, diluted with CH2Cl2 (20 ml), washed with sat. aq. NaHCO3 (2”5 ml) and brine (2”5
ml), dried (Na2SO4), and evaporated under reduced pressure. The residue was purified by CC (silica gel;
cyclohexane/AcOEt/Et3ACHTREGNUN 2 :8 :0.1 (v/v/v)) for 16a and 16b and cyclohexane/AcOEt/Et3ACHTRENUNG (3 :2 :0.05 (v/v/
v) for 16c and 16d).
Phosphoramidite 16a. Yield: 0.23 g (90%). White foam (obtained from 0.2 g (0.3 mmol) of 15a). 1H-
NMR (CD
3A
CN); 3.04–3.15, 3.22–3.33 (2m, 2 Hꢀ
C(3)); 3.35–3.66 (m, OCH2CH2CN, 2 Me
4.34–4.50 (m, 2 HꢀC(1)); 6.65–6.79 (m, 4 arom. H); 7.12–7.22 (m, 5 arom. H); 7.25–7.29 (dd, J=2.2,
9.0, 2 arom. H); 7.39 (m, 2 arom. H); 7.82–7.85 (m, 1 arom. H); 8.04 (dd, J=3.3, 7.9, 1 arom. H); 8.25
(t, J=8.2, 1 arom. H); 8.48–8.54 (m, 2 arom. H). 13C-NMR (CD
3ACHTREUCNG N): 20.79 (CH2CN); 24.77, 24.80,
24.85, 24.90 (4 Me); 43.59, 43.75, 43.90 (C(1), 2 Me2CH); 55.78 (MeO); 59.25 (OCH2CH2CN); 66.55
(C(3)); 71.11 (C(2)); 87.03 (Ar3C); 113.78 (arom. C); 119.16 (CN); 122.91, 123.69, 127.57, 128.58,
128.85, 129.16, 129.40, 130.39, 130.73, 131.09, 131.76, 131.99, 132.55, 133.60, 136.62, 136.87, 145.96,
159.34, 159.42 (arom. C); 163.95, 164.10 (2 C=O). 31P-NMR (CD
CHTRECUNG N): 149.64; 149.57.
3A
Phosphoramidite 16b. Yield: 0.25 g (98%). White foam (obtained from 0.2 g (0.3 mmol) of 15b). 1H-
and 13C-NMR: identical to those of 16a. 31P-NMR (CD
3ACN): 149.61; 149.54.
Phosphoramidite 16c. Yield: 0.60 g (59%). White foam (obtained from 0.75 g (1.3 mmol) of 15c). Rf
(AcOEt/cyclohexane/Et3A
N 2 :3 :0.05) 0.39. 1H-NMR (CDCl3): 0.93, 1.03 (2t, J=6.3, 4 Me); 2.14, 2.38 (2t,
J=7.0, CH2CN); 3.14 (dd, J=5.3, 9.5, HaꢀC(3)); 3.24 (dd, J=5.3, 9.5, HbꢀC(3)); 3.30–3.72 (m,
CTHREUNG
CTHREUNG

Helvetica Chimica Acta – Vol. 89 (2006)
1837
OCH2CH2CN, 2 MeCH); 3.74, 3.76 (2s, 2 MeO); 4.11–4.23 (m, HꢀC(2)); 4.44–4.60 (m, 2 HꢀC(1));
6.69–6.79 (m, 4 arom. H); 7.11–7.48 (m, 9 arom. H); 7.75 (q, J=8.1, 2 arom. H); 8.21 (t, J=8.5, 2
arom. H); 8.57 (t, J=6.3, 2 arom. H). 13C-NMR (CDCl3): 20.06 (CH2CN); 24.11, 24.21, 24.33, 24.45 (4
Me); 42.99 (C(1)); 42.93, 43.17 (2 MeCH); 55.14 (MeO); 57.97 (OCH2CH2CN); 65.62 (C(3)); 70.85
(C(2)); 86.13 (Ar3C); 112.91, 122.92, 126.56, 126.93, 127.68, 128.16, 130.09, 130.91, 131.17, 131.60,
133.73, 136.10, 136.21, 144.75, 158.25, 158.30 (arom. C); 164.13, 164.30 (2 C=O). 31P-NMR (CDCl3):
150.14. EI-MS: 773 (M+).
Phosphoramidite 16d. Yield: 0.12 g (22%). White foam (obtained from 0.40 g (0.7 mmol) of 15d). Rf
(AcOEt/cyclohecane/Et
(CDCl3): 150.14.
N 2 :3 :0.05) 0.39. 1H- and 13C-NMR: identical to those of 16c. 31P-NMR
3ACHTREUNG
General Procedure for N-Alkylation of 19. NaH (60% in mineral oil; 0.07 g, 2.9 mmol) was added
portionwise to the soln. of naphthalimide (19; 0.40 g, 2 mmol) in dry DMF (40 ml). The mixture was stir-
red at r.t. for 30 min, then Bu4ACHTRENUNG Br (TBAB) (0.13 g, 0.4 mmol) was added, and the mixture was stirred for
another 30 min. 2-[(S)-2,2-Dimethyl-1,3-dioxolan-4-yl]ethyl methanesulfonate (17; 0.88 g, 4 mmol) or 3-
[(S)-2,2-dimethyl-1,3-dioxolan-4-yl]propyl methanesulfonate (18; 0.60 g, 2.5 mmol) was added dropwise.
The mixture was stirred for 48 h at 1408. After cooling, DMF was co-evaporated with xylene under
reduced pressure, and the residue was dissolved in H2O and extracted with CHCl3 (3”100 ml). The com-
bined org. phases were dried (MgSO4), and the solvent was removed under reduced pressure. The prod-
uct was purified by CC (silica gel; AcOEt/petroleum ether 2 :3 (v/v)).
2-{2-[(S)-2,2-Dimethyl-1,3-dioxolan-4-yl]ethyl}-1H-benzo[de]isoquinoline-1,3(2H)-dione
(20).
Yield: 0.359 g (54%). Yellow solid. Rf (AcOEt/petroleum ether 2 :3) 0.60. M.p. 1858. 1H-NMR
(CDCl3): 1.33, 1.41 (2s, 2 Me); 1.90–1.99 (m, 1 HꢀC(2)); 2.05–2.15 (m, 1 HꢀC(2)); 3.65 (dd, J=6.0,
7.7, HaꢀC(4’)); 4.12 (dd, J=6.0, 7.7, HbꢀC(4’)); 4.19–4.42 (m, 2 HꢀC(1), HꢀC(3)); 7.75 (t, J=7.7, 2
arom. H); 8.21 (d, J=8.4, 2 arom. H); 8.59 (d, J=6.9, 2 arom. H.). 13C-NMR (CDCl3): 25.61, 26.95 (2
Me); 32.11 (C(2)); 37.36 (C(1)); 69.24 (C(4)); 74.28 (C(3)); 108.91 (Me2C); 122.59, 126.89, 131.16,
133.90 (arom. C.); 164.09 (C=O). HR-MALDI-MS: 348.1217 ([M+Na]+, C
19ACHTRUEGNH19ACHTREUNG
NaNOþ4 ; calc. 348.1206).
(21).
2-{3-[(S)-2,2-Dimethyl-1,3-dioxolan-4-yl]propyl}-1H-benzo[de]isoquinoline-1,3(2H)-dione
Yield: 0.453 g (53%). Yellow oil (obtained from 0.492 g (2.5 mmol) of 19). Rf (AcOEt/petroleum ether
2 :3) 0.35. 1H-NMR (CDCl3): 1.34, 1.39 (2s, 2 Me); 1.65–1.90 (m, 2 HꢀC(2), 2 HꢀC(3)); 3.54 (t,
J=7.5, HꢀC(4’)); 4.11–4.25 (m, 2 HꢀC(1), 2 HꢀC(5’)); 7.75 (t, J=7.4, 2 arom. H); 8.21 (dd, J=1.0,
8.3, 2 arom. H); 8.58 (dd, J=1.0, 7.4, 2 arom. H.). 13C-NMR (CDCl3): 24.45 (C(2)); 25.70, 26.89 (2
Me); 30.90 (C(3)); 40.07 (C(1)); 69.33 (C(5)); 75.71 (C(4)); 108.70 (Me2C); 122.58, 126.87, 128.08,
131.17, 131.52, 133.87 (arom. C.); 164.10 (C=O). HR-MALDI-MS: 362.1349 ([M+Na]+, C NaNO4þ ;
19ACHTERNUGH19ACHTREGUN
calc. 362.1363).
2-[(S)-3,4-Dihydroxybutyl]-1H-benzo[de]isoquinoline-1,3(2H)-dione (10). Compound 20 (0.13 g, 0.4
mmol) was stirred in acetic AcOH/H2O 4 :1 (50 ml) at r.t. overnight. The solvent was co-evaporated two
times with toluene, and the product was purified by CC (silica gel; AcOEt/petroleum ether 3 :7 (v/v)) to
afford 10 (0.1 g, 87%). Yellow solid. Rf (AcOEt/petroleum ether 3 :7) 0.20. M.p. 1608. 1H-NMR (DMSO):
1.55–1.68, 1.78–1.86 (2m, 2 HꢀC(2)); 3.22–3.26 (m, 2 HꢀC(4)); 3.51–3.58 (m, HꢀC(3)); 4.02–4.12 (m,
HaꢀC(1)); 4.20–4.30 (m, HbꢀC(1)); 4.58 (br. s, 2 OH); 7.82–7.88 (m, 2 arom. H); 8.41–8.48 (m, 4 arom.
H.). 13C-NMR (DMSO): 31.94 (C(2)); 37.50 (C(1)); 65.92 (C(4)); 69.88 (C(3)); 122.09, 127.16, 127.29,
130.61, 131.24, 134.18 (arom. C.); 163.34 (C=O). HR-MALDI-MS: 308.0888 ([M+Na]+, C NaNO4þ ;
16ACHTERNUGH15ACHTREGUN
calc. 308.0893).
2-[(S)-4,5-Dihydroxypentyl]-1H-benzo[de]isoquinoline-1,3(2H)-dione (11). Compound 21 (0.359 g,
1 mmol) was stirred in AcOH/H2O 4 :1 (40 ml) at r.t. overnight. The solvent was evaporated under
reduced pressure: 11 (quant.). Yellow oil. 1H-NMR (DMSO): 1.32–1.81 (m, 2 HꢀC(2), 2 HꢀC(3));
3.21–3.33 (m, HꢀC(4), 2 HꢀC(5)); 4.05 (t, J=7.3, 2 HꢀC(1)); 4.46 (br. s, 2 OH); 7.85 (dd, J=7.4,
8.2, 2 arom. H); 8.45 (ddd, J=1.0, 7.4, 11.4, 4 arom. H). 13C-NMR (DMSO): 24.08 (C(2)); 30.95
(C(3)); 39.92 (C(1)); 65.79 (C(5)); 70.98 (C(4)); 122.02, 127.20, 130.70, 131.26, 134.25 (arom. C.);
163.37 (C=O). HR-MALDI-MS: 322.1059 ([M+Na]+, C
2-{(S)-4-[Bis(4-methoxyphenyl)(phenyl)methoxy]-3-hydroxybutyl}-1H-benzo[de]isoquinoline-1,3-
(2H)-dione (22). Compound 10 (0.1 g, 0.34 mmol) was dissolved in dry pyridine (20 ml), and 4,4’-dimeth-
ACHTREUNGoxytrityl chloride (DMT-Cl; 0.15 g, 0.4 mmol) was added portionwise at 08. The mixture was stirred over-
NaNOþ4 ; calc. 322.1050).
17ACHTRUEGNH17ACHTREUNG
ACHTREUNG

1838
Helvetica Chimica Acta – Vol. 89 (2006)
night and co-evaporated with toluene/MeOH 1:1 (20 ml) under reduced pressure. The residue was puri-
fied by CC (silica gel; cyclohexane/AcOEt/Et3AN 49 :49 :2 (v/v/v)) to yield 22 (0.12 g, 60%). Yellow foam.
Rf (cyclohexane/AcOEt/Et3A
CTHREUNG
C
2 HꢀC(3)); 3.30 (br. s, HꢀC(4)); 3.75 (s, 2 MeO); 4.45 (t, J=6.2, 2 HꢀC(1)); 6.77 (d, J=8.8, 4 arom. H.);
7.13 (t, J=7.3, 2 arom. H.); 7.23 (t, J=7.7, 2 arom. H.); 7.29 (dd, J=7.3, 1.9, 4 arom. H); 7.41 (d, J=8.2, 2
arom. H); 7.76 (t, J=7.7, 2 arom. H); 8.22 (d, J=8.2, 2 arom. H); 8.60 (d, J=7.0, 2 arom. H). 13C-NMR
(CDCl3): 32.48 (C(1)); 37.18 (C(2)); 55.15 (2 MeO); 67.41 (C(4)); 68.32 (C(3)); 85.93 (Ar3C); 113.03,
122.46, 126.63, 126.95, 127.72, 128.15, 130.00, 131.42, 134.09, 136.09, 144.84, 158.34 (arom. C.); 164.57
(C=O). HR-MALDI-MS: 610.2220 ([M+Na]+, C
2-{(S)-5-[Bis(4-methoxyphenyl)(phenyl)methoxy]-4-hydroxypentyl}-1H-benzo[de]isoquinoline-1,3-
(2H)-dione (23). Compound 11 (0.300 g, 1.0 mmol) was dissolved in dry CH2Cl2 (20 ml) and dry Et3AN (1
A
ml), and DMT-Cl (0.341 g, 1.0 mmol) was added. The mixture was stirred under N2 overnight and con-
centrated under reduced pressure. The residue was purified by CC (silica gel; AcOEt/petroleum
ether/Et3CAHTRENUNG 24 :74 :2 (v/v/v)) to yield 23 (0.273 g, 45%). Yellow oil. Rf (AcOEt/petroleum ether 1:3)
0.27. 1H-NMR (CDCl3): 1.50–1.61 (m, 2 HꢀC(3)); 1.71–1.94 (m, 2 HꢀC(2)); 3.03 (dd, J=7.5, 9.1,
HaꢀC(5)); 3.15 (dd, J=3.5, 9.1, HbꢀC(5)); 3.77 (s, 2 MeO); 3.82–3.90 (m, HꢀC(4)); 4.20 (t, J=7.4, 2
HꢀC(1)); 6.80 (d, J=9.0, 4 arom. H); 7.18–7.26 (m, 3 arom. H); 7.30 (d, J=9.1, 4 arom. H); 7.41 (d,
J=7.2, 2 arom. H); 7.74 (t, J=8.4, 2 arom. H); 8.22 (dd, J=1.0, 8.4, 2 arom. H); 8.58 (dd, J=1.0, 7.2,
2 arom. H). 13C-NMR (CDCl3): 24.32 (C(2)); 30.68 (C(3)); 40.07 (C(1)); 55.11 (2 MeO); 67.42 (C(5));
70.71 (C(4)); 85.92 (Ar3C); 113.03, 122.58, 126.66, 126.84, 127.72, 128.04, 129.95, 131.12, 131.49, 133.83,
135.93, 144.79, 158.35 (arom. C); 164.11 (C=O). HR-MALDI-MS: 624.2352 ([M+Na]+, C NaNO6þ ;
38ACHTERNUGH35ACHTREGUN
calc. 624.2357).
General Procedure for Synthesis of Phophoramidites 24 and 25. The DMT-protected compound, 22
or 23, was mixed with N,N-diisopropylammonium tetrazolide (1.5 equiv) in dry CH2Cl2 (7 ml). (2-Cya-
noethoxy)bis(diisopropylamino)phosphane (1.5 equiv. for 22, 3 equiv for 23) was added, and the mixture
was stirred under N2 for 24 h. The solvent was evaporated under reduced pressure, and the residue was
purified by CC (silica gel; cyclohexane/AcOEt/Et3ACHTRENUNG 49 : 49 :2 (v/v/v)).
Phosphoramidite 24. Yield: 0.076 g (81%). Yellow foam (obtained from 0.070 g (0.119 mmol) of 22).
1H-NMR (CDCl3): 1.05–1.22 (m, 4 Me); 2.04–2.09 (m, 2 HꢀC(2)), 2.39, 2.69 (2t, CH2CN); 3.05–3.32 (m,
2 HꢀC(4)); 3.53–3.72, 3.84–4.02 (2m, OCH2CH2CN, 2 Me
2ACHTRECUNG H); 3.78 (2s, 2 MeO); 4.13–4.18, 4.31–4.41
(2m, 2 HꢀC(1), HꢀC(3)); 6.78–6.83 (m, 4 arom. H); 7.17–7.29 (m, 3 arom. H); 7.35 (d, J=8.8, 4 arom.
H); 7.47 (dd, J=1.0, 8.4, 2 arom. H); 7.74 (t, J=7.7, 2 arom. H); 8.19 (dd, J=1.0, 8.4, 2 arom. H); 8.58 (d,
J=6.9, 2 arom. H.). 13C-NMR (CDCl3): 20.01 (CH2CN); 24.42, 24.52, 24.62, 24.72 (4 Me); 32.82 (C(2));
37.27 (C(1)); 42.99, 43.08 (2 Me2CH); 55.16 (2 MeO); 58.30 (OCH2CH2CN); 66.19 (C(4)); 72.74 (C(3));
85.98 (Ar3C); 112.99, 122.79, 126.62, 126.84, 127.67, 128.32, 130.15, 131.09, 131.55, 133.77, 136.80, 144.93,
158.36 (arom. C); 163.92 (C=O). 31P-NMR (CHCl3): 149.46; 149.78. HR-MALDI-MS: 810.3309
([M+Na]+, C P+; calc. 810.3278).
46CATHERUNGH50ACHTRUENNG 3ACHTRENUNG aO7ACHTERUNG
Phosphoramidite 25. Yield: 0.141 g (39%). Colorless foam (obtained from 0.273 g (0.45 mmol) of
1
23). Rf (cyclohexane/AcOEt 1:1) 0.57. H-NMR (CDCl3): 0.99–1.15 (m, 4 Me); 1.60–1.93 (m, 2 Hꢀ
C(2), 2 HꢀC(3)); 2.36, 2.60 (2t, J=6.8, CH2CN); 2.94 (dd, J=6.6, 8.9, HaꢀC(5)); 3.08–3.19 (m, Hbꢀ
C(5)); 3.45–3.87 (m, OCH2CH2CN, 2 Me
2A
(m, 2 HꢀC(1)); 6.74–6.79 (m, 4 arom. H); 7.14–7.26 (m, 3 arom. H); 7.29 (dd, J=2.3, 9.1, 4 arom.
H); 7.41 (d, J=7.2, 2 arom. H); 7.74 (t, J=8.4, 2 arom. H); 8.19 (d, J=8.4, 2 arom. H); 8.58 (ddd,
J=1.1, 3.5, 7.2, 2 arom. H). 13C-NMR (CDCl3): 20.22 (CH2CN); 23.45 (C(2)); 24.38, 24.46, 24.60, 24.69
(4 Me); 30.91 (C(3)); 40.28 (C(1)); 42.88, 43.05 (2 Me2CH); 55.14 (2 MeO); 58.45 (OCH2CH2CN);
65.86 (C(5)); 72.95 (C(4)); 85.86 (Ar3C); 112.95, 122.71, 126.58, 126.89, 127.63, 128.15, 130.08, 131.09,
131.55, 133.78, 136.18, 144.99, 158.28 (arom. C); 164.05, 164.09 (2 C=O). 31P-NMR (CDCl3): 149.06;
149.60.
ODN and INA Synthesis, Purification, and Measurement of Melting Temperatures. The ODN and
INA syntheses were carried out on an Expediteꢁ nucleic acid synthesis system model 8909 from Applied
Biosystems. The appropriate naphthalimide phosphoramidite (16a–16d, 24, and 25) was dissolved in a
1:1 mixture of dry MeCN and dry CH2Cl2, as a 0.1M soln., and inserted into the growing oligonucleotide
chain using the same conditions as for normal nucleotide couplings (2-min coupling). The ODNs were

Helvetica Chimica Acta – Vol. 89 (2006)
1839
synthesized with Trityl-on and purified by a Waters Prep LC 4000 HPLC with a Waters Prep LC controller
and a Waters 2487 Dual l absorbance detector on a Waters Xterraꢀ MS C18 column. Buffer A: 950 ml of
0.1M NH4HCO3 and 50 ml of MeCN, (pH 9.0), and buffer B: 250 ml of 0.1 NH4HCO3 and 750 ml of
MeCN, (pH 9.0). Gradients: 5 min 100% A, linear gradient to 70% B in 30 min, 2 min. with 70% B, linear
gradient to 100% B in 8 min and then 100% A in 15 min (product peak at ca. 35 min). The ODNs were
DMT-deprotected in 100 ml of 80% AcOH. After 20 min, 100 ml of H2O, 50 ml of a 3M AcONa, and 600 ml
of abs. EtOH were added. The ODNs were kept at ꢀ208 overnight, by which precipitation occurred.
They were then centrifuged for 15 min upon cooling. The supernatant was decanted off, and the
ODNs were dried in a vacuum centrifuge. All modified ODNs were confirmed by MALDI-TOF analysis
on a Voyager Elite Bio spectrometry research station from Perceptive Biosystems.
Melting-temp. measurements were performed on a Perkin-Elmer Lambda 20 UV/VISspectrometer
fitted with a PTP-6 temp. programmer. Melting temp. (Tm) measurements were determined in a 1 mM
EDTA, 10 mM Na2HPO4 ·2 H2O, 140 mM NaCl buffer at pH 7.0 for 1.5 mM of each strand. (For mismatch
measurements, a concentration of 1.0 mM of each strand was used.) The melting temp. was determined as
the maximum of the first derivative plots of the melting curves obtained by measuring the absorbance at
260 nm against increasing temp. (1.08/min) and is with an uncertainty ꢁ1.08 as determined by repetitive
experiments.
REFERENCES
[1] M. F. BraÇa, A. Ramos, Curr. Med. Chem. – Anti-cancer Agents 2001, 1, 237.
[2] P. F. Bousquet, M. F. BraÇa, D. Conlon, K. M. Fitzgerald, D. Perron, C. Cocchiaro, R. Miller, M.
Moran, J. George, X.-D. Qian, G. Keilhauer, C. A. Romerdahl, Cancer Res. 1995, 55, 1176.
[3] R. J. McRipley, P. E. Burns-Horwitz, P. M. Czerniak, R. J. Diamond, M. A. Diamond, J. L. D. Miller,
R. J. Page, D. L. Dexter, S.-F. Chen, J.-H. Sun, C. H. Behrens, S. P. Seitz, J. L. Gross, Cancer Res.
1994, 54, 159.
[4] M. R. Kirshenbaum, S.-F. Chen, C. H. Behrens, L. M. Papp, M. M. Stafford, J.-H. Sun, D. L. Beh-
rens, J. R. Fredericks, S. T. Polkus, P. Sipple, A. D. Patten, D. Dexter, S. P. Seitz, J. L. Gross, Cancer
Res. 1994, 54, 2199.
[5] J. L. Nitiss, J. Zhou, A. Rose, Y. Hsiung, K. C. Gale, N. Osheroff, Biochemistry 1998, 37, 3078.
[6] C. Bailly, M. BraÇa, M. J. Waring, Eur. J. Chem. 1996, 240, 195.
[7] J. Gallego, B. R. Reid, Biochemistry 1999, 38, 15104.
[8] M. A. Villalona-Calero, J. P. Eder, D. L. Toppmeyer, L. F. Allen, R. Fram, R. Velagapudi, M. Myers,
A. Amato, K. Kagen-Hallet, B. Razvillas, D. W. Kufe, D. D. Von Hoff, E. K. Rowinsky, J. Clin.
Oncol. 2001, 19, 857.
[9] J. Thompson, C. B. Pratt, C. F. Stewart, L. Avery, L. Bowman, W. C. Zamboni, A. Pappo, Invest.
New Drugs 1998, 16, 45.
[10] A. Awada, R. Thçdtmann, M. J. Piccart, J. Wanders, A. H. G. J. Schrijvers, I.-M. Von Broen, A. R.
Hanauske, Eur. J. Cancer 2003, 39, 742.
[11] M. F. BraÇa, M. Cacho, A. Gradillas, B. de Pascual-Teresa, A. Ramos, Curr. Pharm. Des. 2001, 7,
1745.
[12] Y.-H. Hsiang, J. B. Jiang, L. F. Liu, Mol. Pharmacol. 1989, 36, 371.
[13] P. De Isabella, F. Zunino, G. Capranico, Nucleic Acids Res. 1995, 23, 223.
[14] R. Rosell, J. Carles, A. Abad, N. Ribelles, A. Barnadas, A. Benavides, M. Martín, Invest. New Drugs
1992, 10, 171.
[15] H. Ikeda, Y. Nakamura, I. Saito, Tetrahedron Lett. 2002, 43, 5525.
[16] P. E. Nielsen, M. Egholm, R. H. Berg, O. Buchardt, Science, 1991, 254, 1497.
[17] U. B. Christensen, E. B. Pedersen, Nucleic Acids Res. 2002, 30, 4918.
[18] U. B. Christensen, E. B. Pedersen, Helv. Chim. Acta 2003, 86, 2090.
[19] S. Lout, Nature Biotechnol. 23 June 2003, http://dx.doi.org/10.1038/bioent748. INAꢁ is commer-
cially available from Human Genetic Signatures Pty Ltd., Sydney, Australia.
[20] V. V. Filichev, M. H. H. Khalid, U. B. Christensen, E. B. Pedersen, Tetrahedron Lett. 2004, 45, 4907.

1840
Helvetica Chimica Acta – Vol. 89 (2006)
[21] K. Augustyns, A. V. Aerschot, A. V. Schepdael, C. Urbanke, P. Herdewijn, Nucleic Acids Res. 1991,
19, 2587.
[22] K. Mori, M. Ikunaka, Tetrahedron 1984, 40, 3471.
[23] H. Hayashi, K. Nakanishi, C. Brandon, J. Marmur, J. Am. Chem. Soc. 1973, 95, 8749.
[24] H. S. Kim, D. Barak, T. K. Harden, J. L. Boyer, K. A. Jacobson, J. Med. Chem. 2001, 44, 3092.
[25] P. Dostert, M. Langlois, P. Guerret, J.-F. Ancher, B. Bucher, G. Mocquet, Eur J. Med. Chem. 1980,
15, 199.
[26] C. Herdeis, Synthesis 1986, 232.
[27] H. Brunner, H.-J. Lautenschlager, Synthesis 1989, 706.
[28] A. Mujeeb, S. M. Kerwin, G. L. Kenyon, T. L. James, Biochemistry 1993, 32, 13419.
[29] See http://www.schrodinger.com.
[30] W. L. Jørgensen, J. Tirado-Rives, J. Am. Chem. Soc. 1988, 110, 1666.
[31] W. C. Still, A. Tempczyk, R. C. Hawley, T. Hendrickson, J. Am. Chem. Soc. 1990, 112, 6127.
Received July 10, 2006