Novel oligonucleotide analogues containing a morpholinoamidine unit

Article abstract of DOI:10.1016/j.tet.2008.11.069

Morpholinoamidines were devised as new cationic units in oligonucleotides, by combining morpholino-nucleosides (to simplify the nomenclature, we will use the term morpholino-nucleosides to refer to nucleoside analogues in which the ribose ring was transfo

Full text of DOI:10.1016/j.tet.2008.11.069

Tetrahedron 65 (2009) 1171–1179
Contents lists available at ScienceDirect
Tetrahedron
journal homepage: www.elsevier.com/locate/tet
Novel oligonucleotide analogues containing a morpholinoamidine unit
*
´
Sonia Perez-Rentero, Javier Alguacil, Jordi Robles
`
´
´
`
Departament de Qu´ımica Organica, Facultat de Quımica, Universitat de Barcelona, Martı i Franques 1-11, E-08028 Barcelona, Spain
a r t i c l e i n f o
a b s t r a c t
Article history:
Morpholinoamidines were devised as new cationic units in oligonucleotides, by combining morpholino-
nucleosides (to simplify the nomenclature, we will use the term morpholino-nucleosides to refer to nu-
cleoside analogues in which the ribose ring was transformed into a morpholine) with internucleoside
guanidines. Here, methodology was developed to synthesize oligonucleotides containing morpholi-
noamidines formed by morpholino-uridine and 50-amino-50-deoxythymidine. Morpholinoamidine was
produced by solid-phase reaction of Alloc-morpholinocarbothioamide with 50-aminonucleoside resin
and Mukaiyama’s reagent activation. Two 14-mer oligonucleotides containing a single morpholinoami-
dine were synthesized and their affinity properties were investigated by forming DNA double and triple
helices. Duplexes were slightly stabilized by a 30 unit, but were less stable if internally positioned. No-
tably, triplexes were significantly stabilized at pH 7.0.
Received 30 September 2008
Received in revised form 7 November 2008
Accepted 20 November 2008
Available online 27 November 2008
Keywords:
Guanidine
Morpholinoamidine
Oligonucleotide
Nucleic acids analogues
Duplex
Ó 2008 Elsevier Ltd. All rights reserved.
Triplex
1. Introduction
were nuclease-resistant.7d Although several theoretical studies
suggested that GO could be useful in targeting biological se-
As they are endowed with excellent molecular recognition and
self-assembly properties, oligonucleotides have been used in-
creasingly in pharmaceuticals, diagnostics, biotechnology and,
more recently, in nanotechnology.1 However, such a vast range of
applications would be precluded if chemical synthesis had not of-
fered the possibility of introducing modifications into oligonucle-
otide strands,2 to overcome the limitations of the natural backbone,
which are mainly chemical and enzymatic lability, affinity de-
pendent on ionic media and low capacity to penetrate cells. Among
many other analogues, polyarginine oligonucleotide conjugates3
were prepared to show equal or better biophysical and pharma-
codynamic properties than their anionic counterparts. Because
cationic peptides are unstable in serum and are cytotoxic, other
alternatives were explored to introduce cationic groups into oli-
gonucleotides, such as tethering guanidines to phosphoramidate
groups4a or derivatizing nucleobases4b and riboses.4c Guanidine
oligonucleotides (GO) in which the internucleoside phosphate
group is replaced by a guanidine are also known through the work
of Bruice group.5–8 Several types of such oligomers were synthe-
sized6 and showed extremely close affinity to complementary DNA
and RNA sequences,8 essentially due to favorable electrostatic in-
teractions between guanidine and phosphate groups. GO specificity
was not compromised by close affinity, and guanidinium linkages
quences,8 to date no biological assays with guanidine oligonucle-
otides have been reported.
We devised a novel guanidine analogue formed by morpholi-
noamidine units based on the beneficial effects conferred by posi-
tive charges on oligonucleotides (1, Fig. 1). These units resulted
from the combination of two previously known backbones: nu-
cleosides with a morpholine ring such as those forming morpho-
lino oligonucleotides9 (2) and guanidine internucleoside bonds
known in guanidine oligonucleotides (3).
In our view, this new backbone could combine the best of both
morpholino and guanidine analogues. As mentioned, guanidine
groups increased oligonucleotide affinity for target sequences be-
cause of their positive charges and stability to degradation by nu-
cleases. Morpholino oligonucleotides (MO) have promising
properties as antisense agents by a non-RNase H mechanism,9
blocking the production of miRNA10 or altering splicing to correct
B
B
NH
B
O
X
O
O
O
N
N
P
O
HN
H2N
N
O
H2N
NH
NH
(1)
(3)
(2)
* Corresponding author. Tel.: þ34 93 4021253; fax: þ34 93 3397878.
E-mail address: jrobles@ub.edu (J. Robles).
Figure 1. Oligonucleotide analogues containing morpholinoamidine (1), morpholino-
phosphorodiamidate (2) and ribose guanidine units (3).
0040-4020/$ – see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2008.11.069

1172
S. Pe´rez-Rentero et al. / Tetrahedron 65 (2009) 1171–1179
dystrophic pathologies.11 As such, they are very useful in de-
velopmental biology.12 Nonetheless, their main limitation is low
penetration across membranes, but this could be usefully improved
by coadministration13 or by covalent conjugation with polyarginine
peptides.14 Therefore, it seems reasonable that morpholinoamidine
oligonucleotides that contain guanidines immersed in backbone
could also show properties of transfection into cells without being
conjugated with other molecules. With regard to synthetic pro-
cedures, morpholino-nucleosides offer the advantage that they are
obtained from ribonucleosides,15 which are cheaper starting
material than deoxyribonucleosides, and possess one of the amino
functions needed to produce guanidine. Here, we report the
preparation of oligonucleotides containing a single morpholinoa-
midine unit and the preliminary results on their hybridization
properties to form double and triple helices.
DMTO
DMTO
U
O
(6)
N
H
Alloc-NCS
U
O
N
(7)
method A:
method B:
AllocNH
S
in situ activation
preactivation
activator +
DMTO
U
O
N
H2N
O
T
NO2
(8)
(12b)
AllocNH
S
O
NO2
DMTO
2. Results and discussion
H2N
T
O
2.1. Oligonucleotide synthesis
O
N
U
(12b)
O
To obtain morpholinoamidine containing oligonucleotides, we
devised the solid-phase methodology outlined in Scheme 1, which
would permit the preparation of partially or totally modified
strands. Morpholino-nucleosides (4) are used as nucleoside syn-
thons, suitably functionalized to form a guanidine. Initially, we
based ourselves on the solid-phase methodology for preparing GO
described by Bruice,6c–i which employed protected thiourea de-
rivatives (5) as precursors of guanidine function. Following this
strategy, morpholine carbothioamides (thioureas) 5 were produced
from morpholino-nucleosides by reaction with a protected iso-
thiocyanate. At this point, thiourea protection procured activation
of carbothioamide function to nucleophilic addition during guani-
dine formation and, when the guanidine was formed, avoided un-
desired reactions during oligonucleotide synthesis. Guanidine
would be formed by a solid-phase reaction of a preactivated or in
situ-activated morpholinocarbothioamide with an immobilized 50-
aminonucleoside. As shown in the scheme, the method could be
adapted to inserting one single morpholinoamidine unit in the
oligonucleotide strand by using a hydroxy morpholino-nucleoside
(4, X¼O) or adapted to building a morpholinoamidine stretch by
sequential coupling of amine morpholino-nucleosides (4 X¼NH).
Final deprotection and cleavage from the resin give the desired
oligonucleotide.
AllocN
N
O
H
T
(13)
O
i) removal of DMT group
ii) cleavage from resin
i) removal of DMT group
ii) removal of Alloc group
iii) cleavage from resin
HO
HO
O
N
U
O
U
N
AllocN
N
H
T
H2N
N
H
T
O
O
(14a)
(14b)
OH
OH
Scheme 2. Solid-phase synthesis of guanidines 14a and 14b.
synthetic target in order to test the methodology that would permit
the synthesis of more complex molecules. Two methods were
studied to generate guanidine function, as outlined in Scheme 2:
method A used a preactivated isothiourea intermediate (8), and
method B activated thiourea in situ.
Morpholinocarbothioamide 7 was produced as outlined in
Scheme 3. First, morpholino-uridine 6 was obtained from DMT–
uridine by transforming the ribose ring into a morpholine by
adapting a one-step procedure described by Summerton,15 which
consisted of a first treatment of uridine with a mixture of NaIO4 and
NH4HCO3 in aqueous methanol to produce the oxidation of the
20,30-diol to a cyclic aminal intermediate, and in situ reduction to
morpholine by addition of NaBH3CN. Morpholino-uridine 6 was
thus obtained after work-up and purification by column chroma-
tography, in a moderate 60% yield from 5. NMR spectra of mor-
pholino-uridine 6 confirmed the correct stereochemistry of the
Here, we attempted the simplest objective, which consisted of
synthesizing oligonucleotides containing a single morpholinoami-
dine unit formed by morpholino-uridine and 50-amino-50-deoxy-
thymidine. Guanidine 14b (Scheme 2) was chosen as our first
B
B
PG1
X
PG1
X
O
O
N
PG2
N
C
X
S
(X=O, NH)
(5)
N
H
(4)
PG2 NH
B
S
H2N
B
PG1
X
O
O
N
oligonucleotide
N
DMTO
DMTO
U
DMTO
U
U
O
N
O
PG2
N
O
H2N
NH
NH
a)
60%
b)
98%
oligonucleotide
oligonucleotide
only if X= NH
N
H
(6)
HO OH
(7)
Alloc-NH
S
Scheme 1. General synthetic route to obtain oligonucleotides formed by morpholi-
noamidine units.
Scheme 3. Synthesis of morpholinocarbothioamide 7. Reagents and conditions: (a)
NaIO4, NaHCO3, MeOH, 1 hþNaBH3CN, MeOH, 2 h; (b) Alloc-NCS, AcOEt, 3 h.

S. P´erez-Rentero et al. / Tetrahedron 65 (2009) 1171–1179
1173
morpholine ring. Alloc-protected thiourea 7 was prepared by re-
DMTO
DMTO
U
U
O
N
O
N
Sanger's reagent, DBU,
THF, 15 min
action of 6 with Alloc-isothiocyanate16 in a near-quantitative yield.
Our first choice was the Fmoc derivative, after the work of Bruice,6e–i
but in preliminary essays it proved unstable and the origin of sec-
ondary products during guanidine formation. After other attempts,
we finally chose the allyloxycarbonyl group (Alloc), reported else-
where as protecting guanidines in peptides,17 heterocycles,18 and
a dinucleotide guanidine.6d
NO2
85%
Alloc-N
Alloc-NH
S
S
(7)
(8)
NO2
Scheme 5. Preparation of S-2,4-dinitrophenylisothiourea 8.
At this point, we tested the solid-phase synthesis of dinucleo-
sidic guanidine 14 (see Scheme 2). First, a solid support function-
alized with the 50-aminonucleoside (12) was prepared, as outlined
in Scheme 4. 50-MMT-amino-50-deoxythymidine19 (9) was
incorporated on long-chain alkylamine-controlled pore glass
(LCAA-CPG), following the standard methodology via mono-
succinate derivative 10, which was obtained by reaction with suc-
cinic anhydride in pyridine. Before the nucleoside was coupled,
sarcosine was incorporated on solid support, in order to prevent
partial cleavage of succinate linker20 by the base employed to
promote the formation of guanidine. This gave resin 11 with
intermediate 8, so we decided to explore route B (Scheme 2), which
did not require the preparation of a preactivated intermediate.
To carry out the in situ activation of thiourea (Scheme 2), we
chose 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide hydro-
chloride (EDCI) and Mukaiyama’s reagent (MR: 2-chloro-1-methyl-
pyridinium iodide) as activators, two customary reagents in the
solid-phase synthesis of guanidines.22,23 As shown in entry 1.1 of
Table 1, EDCI-mediated coupling of nucleoside was clearly in-
efficient (10% yield). Entries 1.2–1.6 refer to essays by using MR. A
mixture of 3 equiv of nucleoside 7, 4 equiv of triethylamine, and
3 equiv of activator resulted in a moderate 66% coupling yield after
1 h (entry 1.2). By increasing the concentrations of reagents (entry
1.3), the coupling efficiency improved up to 74%. No significant
improvement was obtained by repeating coupling steps or in-
creasing reaction time up to 12 h (entries 1.4–1.7). Product was
cleaved from the resin by treatment at room temperature with
concentrated ammonia for 1 h and analyzed by HPLC. Surprisingly,
two products were observed in chromatograms, the main one
corresponding to the desired guanidine 14a (85%) and a minor one
(15%), which was identified by MALDI-TOF spectrometry as thy-
midine pyridinium salt 15. This product probably formed by SNAr
reaction of Mukaiyama’s activator on 50-amino-50-deoxythymidine
(Scheme 6), a secondary reaction which competed with the acti-
vation of thiourea. Later on, we observed that this secondary
product could be nearly suppressed in the synthesis of other gua-
nidines by reducing the molar excess of activator to 0.8 equiv and
by premixing nucleoside and activator for 90 min before it was
added to the resin.24
a functionalization of 42
moved, and nucleoside 10 was coupled to the resin by a mixture of
diisopropylcarbodiimide (DIP) and catalytic amount of
1-hydroxybenzotriazole (HOBt) in DMF. After coupling, nucleoside
functionalization was determined to be 34 mol/g (resin 12a).
mmol/g. Then, the Fmoc group was re-
a
m
In the current GO solid-phase synthesis, HgCl2 was used to ac-
tivate thiourea function to produce guanidine.6e–i Because of its
toxicity and the production of insoluble HgS during the formation
of guanidine, which should be removed by washings with etha-
nedithiol to prevent clogging of the solid support, we decided to
explore two other methods to activate thiourea.
Method A (Scheme 2) was based on the use of S-2,4-dini-
trophenylisothiourea 8 as a preactivated form of morpholino-
carbothioamide 7. Isothiourea 8 was obtained from thiourea 7 by
reaction with Sanger’s reagent (1-fluoro-2,4-dinitrobenzene) in the
presence of DBU (Scheme 5).21 As its stability is low, 8 was prepared
just before use and purified by aqueous work-up and precipitation
on ether/hexane. Essays to obtain guanidine 14 were performed at
1 mmol scale. First, free amino resin 12b was produced by treatment
with 3% TCA and subsequent neutralization with 5% triethylamine
in CH2Cl2, after which nucleoside was coupled with equimolar
amounts of nucleoside 8 and triethylamine in anhydrous THF.
Coupling proved to be highly inefficient under every condition
assayed, showing 10–41% conversions (data not shown). Coupling
efficiency did not improve significantly by repeating the coupling
steps with freshly prepared reagents, or by increasing neither
coupling time and reagents concentrations from 0.01 M to 0.05 M.
We reasoned that these low yields were caused by instability of
At this point, we assayed obtaining deprotected guanidine 14b
(Scheme 2). After the formation of guanidine 13, the Alloc group
was removed before the cleavage step with ammonia. As summa-
rized in Table 2, different treatments previously reported to pro-
duce the Alloc group removal were assayed, by combination of the
catalyst Pd(PPh3)4 with various scavenger reagents,25–27 in order to
prevent N-allylation, a known secondary reaction in the depro-
tection of Alloc-protected amines.27 After the Alloc removal step,
the crude was cleaved and analyzed by HPLC. No significant dif-
ferences were observed after the different treatments, as summa-
rized in Table 2. Apparently, each treatment produced the total and
T
MMTNH
T
MMTNH
O
O
a)
(10)
OH
Table 1
71%
O
Synthesis of guanidine 14a via an in situ activation
O
HO
(9)
Entry Activator Conditionsa No. couplings Coupling time (h) % Couplingb
O
1.1
1.2
1.3
1.4
1.5
1.6
1.7
EDCI
MR
MR
MR
MR
MR
MR
B1
B2
B3
B3
B3
B4
B4
1
1
1
2
3
1
1
1
1
1
1
1
10
66
74
68
71
76
73
MMTNH
T
O
H
d)
b) c)
N
CONH
H2N
LCAA-CPG
O
(11)
f= 42 µmol/g
1
12
O
O
N
NHCO
a
Reagents and conditions. B1: 7 (3 equiv), EDCI (6 equiv), and triethylamine
(4 equiv) in CH2Cl2; B2: 7 (3 equiv), MR (3 equiv), and TEA (4 equiv) in CH2Cl2; B3: 7
(15 equiv), MR (5 equiv), and triethylamine (20 equiv) in CH2Cl2; B4: 7 (25 equiv),
MR (25 equiv), and TEA (33 equiv) in CH2Cl2.
Coupling yields were determined by the quantification of DMT groups on the
resin and HPLC analyses of crudes after deprotection and cleavage.
f= 34 µmol/g (12a)
Scheme 4. Synthesis of 50-amino-50-deoxynucleosidyl-resin 12a. Reagents and con-
ditions: (a) succinic anhydride, DMAP cat., pyridine, 12 h; (b) N-Fmoc-sarcosine, DIP,
DMF, 2 h; (c) 50% piperidine in DMF, 5 min; (d) 10, DIP, HOBt, DMF, 2 h.
b

1174
S. Pe´rez-Rentero et al. / Tetrahedron 65 (2009) 1171–1179
DMTO
O
N
U
N
CH3
ii) aq. conc. NH3
Cl
H2N
T
N
CH3
N
H
T
i)
O
TEA, DCM
O
AllocN
N
H
T
O
(13)
O
HO
(12b)
(15)
O
Scheme 6. Secondary reaction producing pyridinium salt 15.
Standard oligonucleotide
synthesis
DMT-5'CBzTTTCBzTTCBzTCBzTTmUg
T
(16)
(Alloc)
clean removal of Alloc, with no trace of N-allylated product. Ac-
cordingly, we opted to use dimethyl malonate as scavenger in the
subsequent syntheses.
removal of Alloc group
As we were able to produce a morpholinoamidine derivative
by solid-phase synthesis, we decided to apply this methodology to
the synthesis of oligonucleotides containing a single morpholi-
noamidine. In particular, we synthesized two versions of the same
polypyrimidine strand, one containing a morpholinoamidine at
the 30-end (18b) and another modified at a central position of the
DMT-5'CBzTTTCBzTTCBzTCBzTTmUgT
(17)
deprotection and cleavage
(aq. conc. NH3, 55 ºC, 6 h)
DMT-5'CTTTCTTCTCTTmUgT
(18a)
removal of DMT group
strand (23b). Syntheses are outlined in Schemes
7 and 8,
(80% AcOH, 0 °C, 15 min)
respectively.
5'CTTTCTTCTCTTmUgT
To synthesize oligonucleotide 18b, which contained a 30 mor-
pholinoamidine, we started on the guanidine resin 13 (see Scheme
7). The rest of the oligonucleotide sequence was assembled in the
automatic synthesizer by using the standard phosphoramidite cy-
cle, and keeping the 50-DMT group to ease purification. Alloc group
was first removed by a single treatment with 0.01 M Pd(PPh3)4 and
0.6 M dimethyl malonate in THF for 30 min, and oligonucleotide
was deprotected and cleaved from resin by treatment with aq
concd ammonia at 55 ꢀC for 6 h. DMT-containing fractions were
purified by HPLC and subsequently, the DMT group was removed by
treatment with 80% AcOH. HPLC analyses (Fig. 2) showed three
main bands in crude, which were isolated and characterized by
MALDI-TOF spectrometry. According to MALDI-TOF analyses, the
lowest tR band corresponded to undefined truncated sequences,
probably caused by inefficient capping during the synthesis; the
second band was oligonucleotide 18b (m/z 4082.7 [MꢁH]ꢁ, calcu-
lated mass for C136H183N35O88P12 4085.8) and the highest tR and
main band was due to Alloc-protected oligonucleotide 18c (m/z
4165.9 [MꢁH]ꢁ, calculated mass for C140H187N35O90P12 4169.8). As
the appearance of 18c was due to incomplete removal of the gua-
nidine-protecting group, it was deduced that a single treatment
was insufficient to produce the complete removal of Alloc, most
probably by either sequestration of palladium or steric effects. We
observed that deprotection of guanidine improved when we in-
creased the time of the palladium treatment from 30 min to 1 h
(from 28% to 42% according to the ratio 18b/18c in HPLC profiles)
and carried out repetitive treatments. The best result was obtained
with three treatments of 1 h (86% oligonucleotide in the crude).
(18b)
Scheme 7. Solid-phase synthesis of oligonucleotide 18b, containing a 30 morpholi-
noamidine subunit (mUg refers to the morpholino-uridine amidine unit).
These improvements meant that oligonucleotide 18b could finally
be produced at 13% yield after synthesis and purification (Fig. 2).
The synthetic route to obtain oligonucleotide 23b with a mor-
pholinoamidine unit in a central position of the strand is outlined in
Scheme 8. First, the phosphate segment at the 30 side of morpho-
linoamidine (resin 19) was assembled by standard procedures. To
obtain resin 21, 50-MMT-amino-50-deoxythymidine 30-phosphor-
amidite19 was coupled, the MMT group removed and the guanidine
was formed by coupling morpholinocarbothioamide 7 by activation
with Mukaiyama’s reagent and base (see above). DMT
5'CBzTCBzTTTT
(19)
i) 5'-MMT-amino-5'-deoxythymidine
3'-phosphoramidite, tetrazole
ii) aq. I2
H2N5'
T
O
iii) 2% trichloroacetic acid in CH2Cl2
O
O
CBzTCBzTTTT
(20)
DMTO
O
N
U
i) 13 (25 eq.), MR (25 eq.),
TEA (33 eq.), DCM, 1 h (78%)
ii) Ac2O,
AllocN
N
H
T
O
N-ethyl-N,N-diisopropylamine, DMF
CBzTCBzTTTT
(21)
Table 2
Treatments of Alloc removal to produce guanidine 14b
Standard oligonucleotide
synthesis
Entry
2.1
Treatment
Time
Resulta
0.01 M Pd(PPh3)4,
30 min
Quantitative
deprotection
DMT-5'CBzTTTCBzmUg(Alloc)TCBzTCBzTTTT
(22)
deprotection and cleavage
(aq. conc. NH3, 55 °C, 6 h)
0.6 M dimethyl malonate in THF
0.01 M Pd(PPh3)4,
2.2
10 min
0.6 M trimethylsilylmorpholine, and
0.6 M trimethylsilyl acetate in THF
0.01 M Pd(PPh3)4, 0.6 M HCO2H,
and 0.6 M BuNH2 in THF
0.01 M Pd(PPh3)4, 0.6 M PhSiH3 in
THF
5'
(23a)
DMT- CTTTCmUgTCTCTTTT
2.3
2.4
2.5
1 h
removal of DMT group
(80% AcOH, 0 °C, 15 min)
2ꢂ10 min
2ꢂ10 min
5'CTTTCmUgTCTCTTTT (23b)
0.01 M Pd(PPh3)4, 0.6 M,
BH3$Me2NH in THF
Scheme 8. Solid-phase synthesis of an oligonucleotide (23b), containing a morpholi-
noamidine unit (mUg) in a central position.
a
The efficiency of Alloc removal was determined by HPLC analyses.

S. P´erez-Rentero et al. / Tetrahedron 65 (2009) 1171–1179
1175
determination showed that morpholinoamidine unit formed in
a 78% yield. As guanidine did not form quantitatively, acetylation
was performed to cap the unreacted amino function. The 50 side
sequence was then assembled by standard oligonucleotide pro-
cedures and the 50-DMT group was kept to easy purification (resin
22). The Alloc group was first removed by four treatments with
Pd(PPh3)4 and dimethyl malonate solution (see Table 2). Finally,
oligonucleotide was deprotected and cleaved of the resin by
treatment with aq concd ammonia. Crude, which was more com-
plex than in 18b most probably because of the lower yield of gua-
nidine formation, was purified by HPLC (Fig. 3). The DMT group was
then removed by treatment with 80% AcOH and the resultant
product was analyzed by HPLC to show two main bands (Fig. 3),
which were collected and analyzed by MALDI-TOF spectrometry.
According to mass spectra, the low tR band corresponded to
a mixture of truncated sequences; and the high tR band, to oligo-
nucleotide 23b (m/z 4084.2 [MꢁH]ꢁ). Unlike the synthesis of 18b,
no Alloc-protected oligonucleotide was detected. Oligonucleotide
23b was obtained in 10% yield after synthesis and purification.
Figure 3. HPLC profiles of (a) the crude oligonucleotide 23b after DMT removal and (b)
repurified oligonucleotide 23b. Analyses conditions: linear gradient 10–30%
B in
30 min, where solvent A was 0.05 M AcONH4 and solvent B was CH3CN/water 1:1.
XX
TT
YY
TT
(a) DNA double helices
2.2. UV melting studies
24 25
18b 25
23b 25
'
'
5CTTTCXXCTCTTYY3
TT
mUgT
TT
'
'
3 GAAAGAAGAGAAAA5
mUgT
The affinity properties of oligonucleotides 18b and 23b were
investigated by forming the DNA double and triple helices shown in
Scheme 9, and compared with the unmodified oligonucleotide 24.
For double helices, oligonucleotides 18b, 23b, and 24 were hy-
bridized to polypurine complementary strand 25 at pH 7.0. Melting
curves were recorded at 260 nm and Tm values were determined, to
estimate thermal stability. Since magnesium competes with cat-
ionic ligands for binding to nucleic acids, the effect of its presence
was also investigated, by recording curves with or without 1 mM
MgCl2. Results are summarized in Table 3.
(b) DNA triple helices
XX
TT
YY
TT
'
'
5CTTTCXXCTCTTYY3
24 (26 27)
18b (26 27)
23b (26 27)
'
'
5CTCTGAAAGAAGAGAAAAGTCTC3
TT
mUgT
TT
'
'
3GAGACTTTCTTCTCTTTTCAGAC5
mUgT
Scheme 9. DNA double and triple helices studied. Segments of strands 26 and 27
targeted to form triple helix are shown in a dashed box.
of morpholine-amidine, not compensated by electrostatic in-
teractions with phosphates. In a similar context, a single UgU was
According to the Tm values, modified oligonucleotides 18b and
23b formed duplexes of differing stability. The 30 modified strand
18b formed a slightly more stable duplex than unmodified 24
reported to produce a slight increase in Tm
(D
Tm¼0.3–0.6 ꢀC).8a We
reasoned that because of the planarity of the guanidinium group and
the extended structure of the morpholine ring, the introduction of
a morpholinoamidine in a central position had a disturbing effect on
the regularity of duplex.
(
D
Tm¼þ1.5 ꢀC). Therefore, a single guanidine-positive charge at the
end of the strand 18b improved the affinity to complementary
strand, probably because the morpholinoamidine unity had a co-
hesive effect on partially hybridized edges, by favorable electrostatic
interactions between the guanidinium group and phosphates.
According to electrostaticeffects, the Tm decreasedin thepresence of
Strands 18b and 23b were also assayed as triple helix-forming
oligonucleotides, by hybridization with 23-mer duplex 26$27,
which contains a 14-mer stretch targetable by the parallel motif.
This type of triplex is formed by T(A.T) and Cþ(G.C) triads and with
a parallel orientation of one polypyrimidine strands (called the
Hoogsteen or third stand) with respect to the central polypurine.28
As third-strand cytosines need to be protonated to form Cþ(G.C)
triads, these triplexes are usually stable at pH <7.29 Because of that,
UV melting experiments were carried out at three different pHs.
MgCl2 was incorporated in buffers to procure the formation of
1 mM MgCl2 (
D
Tm¼þ1.0 ꢀC) because of the sheltering of phosphates
by magnesium. The increase in Tm produced by 18b was consistent
with the effect produced for an oligonucleotide containing a single
UgU unit at the 30 end of a duplex (
trary, strand 23b with the morpholinoamidine subunit in a central
position formed a less stable duplex (
D
Tm¼0.7–1.5 ꢀC).8a On the con-
D
Tm¼ꢁ7.5/ꢁ8.0 ꢀC). Here, the
destabilization of duplex was most probably due to steric hindrance
Figure 2. HPLC profiles of (a) crude oligonucleotide 18b, (b) repurified oligonucleotide 18b and (c) Alloc-protected oligonucleotide 18c. Analyses conditions: linear gradient 10–30%
B in 30 min, where solvent A was 0.05 M AcONH4 and solvent B was CH3CN/water 1:1.

1176
S. Pe´rez-Rentero et al. / Tetrahedron 65 (2009) 1171–1179
Table 3
Tm values of duplexes
interactions between guanidine and phosphate explained the for-
mation of triplex under these conditions. Nonetheless, the stabili-
zation was lower than for 18b, probably because of the steric
hindrance produced by the internal morpholinoamidine. At pH 6.5,
triplex increased Tm1 up to 21.5 ꢀC, but it was less stable than un-
a
b
Duplex
MgCl2 (mM)
Tm (ꢀC)
D
Tm (ꢀC)
24$25
d
d
d
1
1
1
42.5
44.0
34.5
42.5
43.5
34.0
d
18b$25
23b$25
24$25
18b$25
23b$25
þ1.5
ꢁ8.0
d
modified triplex (
D
Tm1¼ꢁ5.5 ꢀC). Notably, the triplex formed by
23b at pH 6.0 was not only more stable than at pH 6.5, due to in-
creased protonation of cytosines, but was a bit more stable
(Tm1¼38.0 ꢀC) than the complex formed by 18b (Tm1¼36.0 ꢀC).
þ1.0
ꢁ7.5
a
Experiments were performed with oligonucleotides at 1.5
m
M concentration, in
40 mM NaH2PO4/Na2HPO4 and 140 mM NaCl buffer, with or without 1 mM MgCl2, at
pH 7.0.
3. Conclusions
b
Differences between Tm values of modified and unmodified duplexes.
Morpholinoamidines were devised as new cationic units in ol-
igonucleotides, by combining morpholino-nucleosides with inter-
nucleoside guanidines. Here, methodology was developed to
accomplish the simplest of objectives, the synthesis of oligonucle-
otides containing a single morpholinoamidine unit formed by
morpholino-uridine and 50-amino-50-deoxythymidine. Morpholi-
noamidine could be produced on solid-phase, by reaction of an
Alloc-morpholinocarbothioamide uridine derivative with a 50-
amino-50-deoxythymidine bound to resin, by in situ activation with
Mukaiyama’s reagent. When the oligonucleotide had been assem-
bled, our procedure followed with the removal of the Alloc group
from the guanidine by treatment with Pd0, and subsequent
deprotection and cleavage from the resin by aq concd ammonia.
Following this methodology, two oligonucleotides containing one
morpholinoamidine subunit were synthesized, the one located at
the 30 end, and the other in the middle of the strand. We in-
vestigated the affinity properties of these two oligonucleotides by
forming double and triple helices with DNA complementary
strands. Double helices were found to be slightly stabilized by a 30
morpholinoamidine, but complexes were less stable when located
in the middle of the sequence. Notably, triple helices were signifi-
cantly stabilized at pH 7.0 by the presence of morpholinoamidine
units, whereas under the same conditions unmodified strands did
not produce triplex. According to our results, the stabilization effect
of morpholinoamidine was probably produced by favorable elec-
trostatic interactions between the cationic guanidinium and
phosphates, whereas duplex destabilization when morpholinoa-
midine was internally positioned could originate in steric hin-
drance. In the light of these preliminary results, which revealed
that a single morpholinoamidine unit can improve the affinity of
oligonucleotides, we are working on optimizing and extending the
synthetic methodology, to prepare further modified oligonucleo-
tides and, gain better understanding of their properties.
triple helices.30 When triple helices were formed, melting curves
showed two transitions. The transition at higher temperature was
due to duplex 26$27 dissociation. Since the low temperature
transition was caused by third-strand dissociation (24, 18b, or 23b),
the corresponding melting temperature (Tm1) is a measurement of
the thermal stability of triplex. Table 4 summarizes the Tm1 values
of our experiments.
As expected, the stability of the triple helix formed by un-
modified strand 24 was strongly dependent on pH. Triplexes were
observed at pH 6.0 and 6.5, but not at pH 7.0. Because of increased
protonation of cytosine, triplex proved more stable at pH 6.0
(Tm1¼31 ꢀC) than at pH 6.5 (Tm1¼27 ꢀC). Notably, triplexes formed
by the 30 modified strand 18b were observed at the three different
pHs. At pH 7.0, the triplex was formed with a Tm1 of 24.0 ꢀC,
whereas under the same conditions, triplex from unmodified
strand 24 was not detected. Therefore, a single morpholinoamidine
caused a significant increase in the stability of triplex at pH 7.0. To
explain these results, we speculated on the presence of the gua-
nidinium group decreased electrostatic repulsion between phos-
phate strands and compensated for the loss of hydrogen bonds and
stacking interactions by the partially protonated third-strand cy-
tosines at higher pH. This was the main reasoning to account for the
extremely high stability of triplexes formed by GO.7 The stabilizing
effect was also observed at lower pHs, with
D
Tm¼þ5 ꢀC over the
unmodified triplex. In summary, at every pH studied, the triplex
was stabilized by a single 30 morpholinoamidine subunit in the
third strand. Triplex formed by strand 23b behaved differently from
18b. Triplex was observed at pH 7.0, although with lower stability
than for 18b (Tm1¼13.0 ꢀC). Therefore, a single morpholinoamidine
unit also significantly increased stability at pH 7.0 when internally
positioned. Similar to the effect of 18b, favorable electrostatic
4. Experimental
4.1. General
Table 4
Tm values of triplexes
Triplexa
pH
Tm1 (ꢀC)
D
Tm1 (ꢀC)
b
c
Unless otherwise indicated, all chemicals were purchased from
commercial suppliers (reagent grade) and used without purifica-
tion. Dry CH3CN was obtained by distillation over CaH2 and storage
over CaH2 lumps. CH2Cl2 was neutralized and dried by passing
through basic Al2O3 and storage over CaH2. Dry THF was obtained
by distillation over sodium metal in the presence of benzophenone.
DMF was bubbled with nitrogen to remove volatile contaminants
and dried by storage over CaH2. Amine-free pyridine was obtained
by distillation over ninhydrin and further dried by storage over
CaH2 lumps. NMR spectra were recorded with either a Varian Unity
300 MHz or a Varian Mercury 400 MHz, respectively. UV analyses
and melting curves were performed in a Jasco V-550 instrument
equipped with a thermoregulated cell holder. Low resolution
electrospray mass (EIMS) spectra of nucleosides were recorded in
a Waters Alliance 2696 HPLC coupled to Waters Micromass ZQ MS
detector. MALDI-TOF spectra of oligonucleotides were recorded in
24(26$27)
7.0
6.5
6.0
7.0
6.5
6.0
7.0
6.5
6.0
ndd
d
27.0
31.0
24.0
32.0
36.5
13.0
21.5
38.0
d
d
de
18b(26$27)
23b(26$27)
þ5.0
þ5.5
de
ꢁ5.5
þ7.0
a
Experiments were performed with oligonucleotides at 1
m
M concentration, in
40 mM NaH2PO4/Na2HPO4, 140 mM NaCl and 1 mM MgCl2 buffer, at the pH
indicated.
b
Tm1 referred to the triple helix dissociation.
Differences between Tm1 values of modified and unmodified triplexes at the
c
same pH.
d
nd: not detected.
e
DTm1 could not be calculated because triplex 24(26$27) was no detected at pH
7.0.

S. P´erez-Rentero et al. / Tetrahedron 65 (2009) 1171–1179
1177
a Perseptive Biosystems Voyager DE-RP instrument, by using 2,4,6-
trihydroxyacetophenone (THAP), 2,4,6-trihydroxyacetophenone/
ammonium citrate (1:1) (THAP-CA) or alpha-cyano-4-hydroxycin-
namic acid (ACH) as matrixes.
was suspended in DMF and reacted for 30 min, the resin was
washed with DMF and the aminoacid coupling was repeated, to
render an aminoacid functionalization of 42 mmol/g. The resin was
treated with a solution of Ac2O/DMF 1:1 to cap free amino groups
on the resin. Subsequently, free amine resin 11 was produced by
removal of the Fmoc group with 50% piperidine in DMF and after
this, 50-MMT-amino-50-deoxythymidine 30-monosuccinate (10,
4.2. Synthesis of guanidines 14a and 14b
4.2.1. Morpholino-uridine (6)
63 mmol), HOBt (63 mmol), and diisopropylcarbodiimide (336 mmol)
DMT-U (3.5 mmol) was dissolved in MeOH (100 mL), NaIO4
(3.8 mmol) and NH4HCO3 (7.0 mmol) were added, and the mixture
was left to react for 1 h. Then, the solid that formed was separated
by filtration through glass wool, NaBH3CN (8.7 mmol) was added to
the filtrate and left to react for additional 2 h. After that, the solu-
tion was acidified to pH 5.0 by addition of small portions of 10%
citric acid, and evaporated to dryness. The residue was dissolved in
AcOEt, and the corresponding organic phase was washed with
water (50 mL) and brine, dried over MgSO4, and the solvent was
evaporated. Compound 6 was obtained as a white solid after
purification by silica gel column chromatography, by elution
with CH2Cl2/triethylamine 99:1 and increasing amounts of MeOH
(up to 10%).
were added to the resin and left to react for 2 h. After washings, the
resin was treated with Ac2O/DMF 1:1. The incorporation of nucle-
oside was determined by MMT quantification to be 34 mmol/g.
4.2.4. Guanidine 14a (mUg(Alloc)T)
Resin 12a (1 mmol) was placed into a cylindric polypropylene
reactor normally used for the synthesis of oligonucleotides, and by
the use of two syringes adapted to the reactor it was treated with
3% TCA in CH2Cl2 to produce the removal of MMT group, and
washed with DMF and 5% triethylamine in CH2Cl2 to neutralize the
amine. Under the best conditions (see Table 1), coupling was car-
ried out by mixing approx. 150
mL of a solution 0.15 M of morpho-
linocarbothioamide 7 and 0.23 M of triethylamine and 150
mL of
Yield: 60%. mp: 111–115 ꢀC. Rf (dichloromethane/MeOH 10:1):
a solution 0.15 M of Mukaiyama’s reagent, both in CH2Cl2 and left to
react for 1 h. After this, the resin was washed with DMF and CH2Cl2
and the DMT group was removed with 3% trichloroacetic acid in
CH2Cl2. In this case, the yield of guanidine formation was de-
termined by UV quantification of DMTþ solution (76%). The resin
was treated with aq concd ammonia for 30 min to produce the
cleavage. The crude was analyzed by HPLC to show guanidine 14b
(85%) and pyridium salt 15 (15%). Finally, both products were pu-
rified by HPLC and characterized by mass spectrometry.
mUg(Alloc)T (14a): 32% yield. Reversed-phase HPLC (Kromasil-
C18, linear gradient 5–100% B in 30 min, solvent A: 0.05 M AcONH4,
solvent B CH3CN/water 1:1) tR: 7.9 min. MALDI-TOF MS m/z (posi-
tive mode, ACH, [MþH]þ) calcd for C24H32N7O10 577.2, found 578.0.
Pyridinium salt (15): 5% yield. Reversed-phase HPLC (Kromasil-
C18, linear gradient 5–100% B in 30 min, solvent A: 0.05 M AcONH4,
solvent B CH3CN/water 1:1) tR: 5.6 min. MALDI-TOF MS m/z (posi-
tive mode, ACH, [M]þ) calcd for C16H21N4O4 333.2, found 333.8.
0.4. 1H NMR (300 MHz, CDCl3)
d
(ppm): 7.45 (1H, d, J¼7.6 Hz), 7.33–
7.21 (9H, m), 6.82 (4H, d, J¼8.8 Hz), 5.77 (1H, d, J¼7.6 Hz), 5.68 (1H,
dd, J¼9.7, 2.4 Hz), 3.96 (1H, m), 3.78 (6H, s), 3.24 (1H, dd, J¼9.5,
4.9 Hz), 3.15 (1H, dd, J¼12.1, 2.4 Hz), 3.07 (1H, dd, J¼9.5, 5.1 Hz),
3.01 (1H, dd, J¼12.7, 2.3 Hz), 2.58 (1H, dd, J¼12.7, 10.7 Hz), 2.54 (1H,
dd, J¼12.1, 9.7 Hz). 13C NMR (75 MHz, CDCl3)
d (ppm): 162.7, 158.5,
150.0, 144.2, 139.8, 135.7, 130.0, 128.0, 127.7, 126.3, 113.0, 102.0, 86.0,
80.0, 64.0, 55.1, 49.1, 46.4. EIMS m/z (positive mode, [MþH]þ) calcd
for C30H32N3O6 530.2, found 530.3; HRMS (positive mode,
[MþNa]þ) calcd for C30H31N3O6Na 552.2105, found 552.2098.
4.2.2. Morpholino-uridine carbothioamide (7)
Allyloxycarbonyl chloride (2 mmol) was dissolved in anhydrous
AcOEt (5 mL) and added dropwise on a suspension of potassium
thiocyanate (2.4 mmol) in anhydrous AcOEt (5 mL), and the mix-
ture was left to react for 3 h at room temperature. After that, a so-
lution of nucleoside 6 in anhydrous AcOEt and under argon was
prepared and added dropwise on the previous mixture, and left to
react for additional 2 h. After this, the solvent was removed by
evaporation, the residue was redissolved in CH2Cl2 and the re-
sultant organic phase was washed twice with 10% NaHCO3 and
brine, dried over MgSO4, and evaporated to dryness. Compound 7
was obtained as a white solid, which does not need further puri-
fication according to TLC and NMR analyses.
4.2.5. Guanidine 14b (mUgT)
Guanidine was produced with the same procedure as for 14a,
with an additional deprotection step to remove the Alloc group
before the cleavage step with ammonia. As a general procedure (see
Table 2 for reagents), the resin contained in one synthesis reactor
was vacuum dried, purged with argon, and the reactor was capped
with two all plastic syringes. Then, 250
mL of a solution 1.2 M of
Yield: 98%. mp: 80–82 ꢀC. Rf (dichloromethane/MeOH/triethyl-
scavenger and 250 L of a solution 0.02 M of Pd(PPh3)4 in anhy-
m
amine 10:0.2:0.1): 0.3. 1H NMR (400 MHz, CDCl3)
d
(ppm): 7.42 (1H,
drous THF were delivered into the reactor, and left to react for the
time indicated in Table 2. After this, the resin was filtered and
washed with THF, acetone, 0.1 M sodium diethyldithiocarbamate,
acetone, and water to remove the traces of palladium adsorbed on
the resin. Finally, the resin was treated with aq concd ammonia for
30 min to produce cleavage, and the corresponding crude was
purified by HPLC.
mUgT 14b: 29% yield. Reversed-phase HPLC (Kromasil-C18, linear
gradient 5–100% B in 30 min, solvent A: 0.05 M AcONH4, solvent B
CH3CN/water 1:1) tR: 5.4 min. MALDI-TOF MS m/z (positive mode,
ACH, [MþH]þ) calcd for C20H28N7O8 494.2, found 495.5.
d, J¼7.6 Hz), 7.31–7.24 (9H, m), 6.82 (4H, d, J¼8.8 Hz), 5.92–5.82
(2H, m), 579 (1H, d, J¼7.6 Hz), 530 (1H, d, J¼17 Hz), 5.22 (1H, d,
J¼8.8 Hz), 4.63 (1H, dd, J¼12, 4.4 Hz), 4.57 (1H, dd, J¼12.2, 4.8 Hz),
4.15–4.12 (1H, m), 3.77 (6H, s), 3.29–3.08 (6H, m). 13C NMR
(100 MHz, CDCl3)
d (ppm): 178.4, 163.6, 160.2, 150.1, 149.9, 144.3,
139.6, 135.4, 131.2, 129.7, 128.5, 127.6, 126.2, 119.2, 113.3, 102.2, 85.7,
80.2, 67.5, 64.5, 55.2, 49.7. EIMS m/z (positive mode, [MþH]þ) calcd
for C35H37N4O8S 672.2, found 672.1. HRMS (positive mode,
[MþNa]þ) calcd for C35H36N4O8SNa 695.2146, found 695.2145.
4.2.3. 50-MMT-amino-50-deoxythymidin-3-yl resin (12a)
1 g of commercial LCAA-CPG (Sigma, 500 Å, approx. 50
mmol a-
4.3. Oligonucleotide synthesis
mino/g) was placed into a polypropylene syringe fitted with
a polyethylene disc and sequentially washed with CH2Cl2, THF, 10%
HCl, THF, DMF, 20% piperidine in DMF, 2% TCA in CH2Cl2, 2% N-
ethyl-N,N-diisopropylamine in CH2Cl2, CH2Cl2, CH3OH, and CH3CN.
Unmodified oligonucleotides 24–27 and the phosphate frag-
ments of modified oligonucleotides 18b and 23b were assembled
on CPG in an automatic synthesizer (ABI 380B or Expedite) using
the standard phosphite triester methodology. For the synthesis of
oligonucleotide 18b, resin 13 was first prepared as described above,
N-Fmoc-sarcosine (Novabiochem, 50
mmol) and diisopropylcarbo-
diimide (50 mol) were added to the resin, the resulting mixture
m

1178
S. Pe´rez-Rentero et al. / Tetrahedron 65 (2009) 1171–1179
and after the guanidine formation, the phosphate stretch was as-
sembled using the standard phosphoramidite methodology. For the
synthesis of 23b, 50-MMT-amino-50-deoxythymidine 30-phosphor-
amidite21 was incorporated during the oligonucleotide synthesis in
the same conditions than the rest of phosphoramidites. Then, the
coupling of morpholinocarbothioamide 7 was performed as de-
scribed for the synthesis of guanidine 14a. Once the guanidine was
formed, the resin was returned to the synthesizer and the 50
phosphate stretch was assembled using the standard protocol. Ol-
igonucleotides 24–27 were finally deprotected and cleaved from
the resin by treatment with aq concd ammonia at 60 ꢀC (6–12 h),
and purified as explained below. Oligonucleotides 18b and 23b
were deprotected by a two-step procedure. In the first place, the
Supplementary data
1H and 13C NMR spectra and mass spectra of morpholino-uri-
dine 6 and morpholino-uridine carbothioamide 7, and UV melting
profiles of duplexes and triplexes summarized in Tables 3 and 4.
Supplementary data associated with this article can be found in the
online version, at doi:10.1016/j.tet.2008.11.069.
References and notes
1. (a) Crooke, S. T. Ann. Rev. Med. 2004, 55, 61–95; (b) Kurreck, J. In Therapeutic
Oligonucleotides; Kurreck, J., Ed.; Royal Society of Chemistry: Cambridge, 2008;
pp 1–22; (c) Tachikawa, K.; Briggs, S. P. Curr. Opin. Biotechnol. 2006, 17, 659–665;
(d) Dubertert, B. Nat. Mater. 2005, 4, 797–798; (e) Seeman, N. C. Nature 2003,
421, 427–431.
2. (a) Wilson, C.; Keefe, A. D. Curr. Opin. Chem. Biol. 2006, 10, 607–614; (b) Bum-
crot, D.; Manoharan, M.; Koteliansky, V.; Sah, D. W. Y. Nat. Chem. Biol. 2006, 2,
711–719.
oligonucleotide resin (1
was submitted under argon to three treatments with 250
1.2 M dimethyl malonate and 250 L of 0.02 M Pd(PPh3)4 in anhy-
mmol) was placed into a fritted syringe and
mL of
m
drous THF for 1 h. To produce the complete deprotection and the
cleavage of the resin, the resin was placed into a vial, and treated
with aq concd ammonia at 60 ꢀC for 6 h. DMT–oligonucleotides
were purified by reversed-phase HPLC (Hamilton PRP, linear gra-
dients 10–50 or 60% B in 30 min). The collected fractions were
lyophylized and treated with AcOH/H2O 4:1 (15 min, 0 ꢀC) to pro-
duce the removal of DMT and lyophylized. Oligonucleotides 18b
and 23b needed to be repurified by reversed-phase HPLC (Hamilton
PRP, linear gradients 10–30% B in 30 min). Oligonucleotides were
quantified by UV absorption at 260 nm. 3260 values calculated by
the nearest-neighbor method.
Oligonucleotide 18b: 13% yield. Reversed-phase HPLC (Kromasil-
C18, linear gradient 10–30% B in 30 min, solvent A: 0.05 M AcONH4,
solvent B CH3CN/water 1:1) tR: 17.8 min. MALDI-TOF MS m/z
(negative mode, THAP-CA, [MꢁH]ꢁ) calcd for C136H182N35O88P12
4085.8, found 4082.7.
3. (a) Fraley, A. W.; Pons, B.; Dalkara, D.; Nullans, G.; Behr, J.-P.; Zuber, G. J. Am.
Chem. Soc. 2006, 128, 10763–10771; (b) Zatsepin, T. S.; Turner, J. J.; Oretskaya, T.
S.; Gait, M. J. Curr. Pharm. Design 2005, 11, 3639–3654.
4. (a) Ohmichi, T.; Kuwahara, M.; Sasaki, N.; Hasegawa, M.; Nishikata, T.; Sawai,
H.; Sugimoto, N. Angew. Chem., Int. Ed. 2005, 44, 6682–6685; (b) Deglane, G.;
´
Abes, S.; Michel, T.; Prevot, P.; Vives, E.; Debart, F.; Barvik, I.; Lebleu, B.; Vasseur,
J.-J. ChemBioChem 2006, 7, 684–692; (c) Prakash, T. P.; Pu¨schl, A.; Lesnik, E.;
Mohan, V.; Tereshko, V.; Egli, M.; Manoharan, M. Org. Lett. 2004, 6, 1971–1974.
5. Dempcy, R. O.; Almarsson, O.; Bruice, T. C. Proc. Natl. Acad. Sci. U.S.A. 1994, 91,
7864–7868.
6. (a) Dempcy, R. O.; Browne, K. A.; Bruice, T. C. Proc. Natl. Acad. Sci. U.S.A. 1995, 92,
6097–6101; (b) Dempcy, R. O.; Luo, J.; Bruice, T. C. Proc. Natl. Acad. Sci. U.S.A.
1996, 93, 4326–4330; (c) Linkletter, B. A.; Bruice, T. C. Bioorg. Med. Chem. Lett.
1998, 8, 1285–1290; (d) Barawkar, D. A.; Linkletter, B.; Bruice, T. C. Bioorg. Med.
Chem. Lett. 1998, 8, 1517–1520; (e) Kojima, N.; Bruice, T. C. Org. Lett. 2000, 2, 81–
84; (f) Linkletter, B. A.; Szabo, I. E.; Bruice, T. C. Nucleic Acids Res. 2001, 29, 2370–
2376; (g) Reddy, P. M.; Bruice, T. C. Bioorg. Med. Chem. Lett. 2003, 13, 1281–1285;
(h) Challa, H.; Bruice, T. C. Bioorg. Med. Chem. 2004, 12, 1475–1481; (i) Jain, M. L.;
Bruice, T. C. Bioorg. Med. Chem. 2006, 14, 7333–7346.
7. (a) Browne, K. A.; Dempcy, R. O.; Bruice, T. C. Proc. Natl. Acad. Sci. U.S.A. 1995, 92,
7051–7055; (b) Blasko´ , A.; Dempcy, R. O.; Minyat, E. E.; Bruice, T. C. J. Am. Chem.
Soc. 1996, 118, 7892–7899; (c) Blasko´, A.; Minyat, E. E.; Dempcy, R. O.; Bruice, T.
C. Biochemistry 1997, 36, 7821–7831; (d) Barawkar, D. A.; Bruice, T. C. Proc. Natl.
Acad. Sci. U.S.A. 1998, 95, 11047–11052; (e) Linkletter, B. A.; Szabo, I. E.; Bruice, T.
C. J. Am. Chem. Soc. 1999, 121, 3888–3896; (f) Szabo, I. E.; Bruice, T. C. Bioorg.
Med. Chem. 2004, 12, 4233–4244; (g) Park, M.; Bruice, T. C. Bioorg. Med. Chem.
Lett. 2005, 15, 3247–3251; (h) Toporowski, J. W.; Reddy, S. Y.; Bruice, T. C. Bioorg.
Med. Chem. 2005, 13, 3691–3698; (i) Park, M.; Toporowski, J. W.; Bruice, T. C.
Bioorg. Med. Chem. 2006, 14, 1743–1749; (j) Toporowski, J. W.; Reddy, S. Y.;
Bruice, T. C. Biophys. Chem. 2007, 126, 132–139.
Oligonucleotide 23b: 10% yield. Reversed-phase HPLC (Kromasil-
C18, linear gradient 10–30% B in 30 min, solvent A: 0.05 M AcONH4,
solvent B CH3CN/water 1:1) tR: 17.1 min. MALDI-TOF MS m/z
(negative mode, THAP-CA, [MꢁH]ꢁ) calcd for C136H182N35O88P12
4085.8, found 4084.2.
4.4. UV melting curves
8. (a) Park, M.; Canzio, D.; Bruice, T. C. Bioorg. Med. Chem. Lett. 2008, 18, 2377–
2384; (b) Park, M.; Bruice, T. C. Bioorg. Med. Chem. Lett. 2008, 18, 3488–3491; (c)
Zhang, X.; Bruice, T. C. Bioorg. Med. Chem. Lett. 2008, 18, 665–669.
9. (a) Summerton, J.; Weller, D. Antisense Nucleic Acid Drug Dev. 1997, 7, 187–195;
(b) Summerton, J. Biochim. Biophys. Acta 1999, 1489, 141–158.
10. Dore, L. C.; Amigo, J. D.; dos Santos, C. O.; Zhang, Z.; Gai, X.; Tobias, J. W.; Yu, D.;
Klein, A. M.; Dorman, C.; Wu, W.; Hardison, R. C.; Paw, B. H.; Weiss, M. J. Proc.
Natl. Acad. Sci. U.S.A. 2008, 105, 3333–3338.
11. Fletcher, S.; Honeyman, K.; Fall, A. M.; Harding, P. L.; Johnsen, R. D.; Steinhaus, J.
P.; Moulton, H. M.; Iversen, P. L.; Wilton, S. D. Mol. Ther. 2007, 15, 1587–1592.
12. Corey, D. R.; Abrams, J. M. Genome Biol. 2001, 2, 1–3.
13. (a) Moulton, H. M.; Hase, M. C.; Smith, K. M.; Iversen, P. L. Antisense Nucleic Acid
Drug Dev. 2003, 13, 31–43; (b) Wu, R. P.; Youngblood, D. S.; Hassinger, J. N.;
Lovejoy, C. E.; Nelson, M. H.; Iversen, P. L.; Moulton, H. M. Nucleic Acids Res.
2007, 35, 5182–5191.
14. (a) Moulton, H. M.; Nelson, M. H.; Hatlevig, S. A.; Reddy, M. T.; Iversen, P. L.
Bioconjugate Chem. 2004, 15, 290–299; (b) Nelson, M. H.; Stein, D. A.; Kroeker,
A. D.; Hatlevig, S. A.; Iversen, P. L.; Moulton, H. M. Bioconjugate Chem. 2005, 16,
959–966; (c) Youngblood, D. S.; Hatlevig, S. A.; Hassinger, J. N.; Iversen, P. L.;
Moulton, H. M. Bioconjugate Chem. 2007, 18, 50–60; (d) Amantana, A.; Moulton,
H. M.; Cate, M. L.; Reddy, M. T.; Whitehead, T.; Hassinger, J. N.; Youngblood, D.
S.; Iversen, P. L. Bioconjugate Chem. 2007, 18, 1325–1331; (e) He, J.; Liu, G.; Dou,
S.; Gupta, S.; Rusckowski, M.; Hnatowich, D. Bioconjugate Chem. 2007, 18, 983–
988.
Experiments were performed with oligonucleotides at 1.5
mM
concentration for duplexes and 1 M for triplexes, in 40 mM
m
NaH2PO4/Na2HPO4 and 140 mM NaCl buffer, with or without 1 mM
MgCl2, at pH 7.0 for duplexes, and with 1 mM MgCl2 at pH 6.0, 6.5,
or 7.0 in the case of triplexes. Samples were heated at 90 ꢀC for
5 min, allowed to cool down slowly to room temperature to induce
annealing, and then kept overnight in a refrigerator (5 ꢀC). Melting
curves were recorded by heating the samples from 5 to 90 ꢀC at
a constant rate of 0.5 ꢀC/min and monitoring the absorbance at
260 nm at a sampling rate of 6 points/min. At temperatures below
25 ꢀC, nitrogen was flushed to prevent water condensation on cu-
vettes. Tm values were determined from the maxima of the first
derivative of the curves, calculated with the Microcal Origin soft-
ware. Experiments were repeated until coincident Tm values were
obtained. The error in Tm values was estimated to be ꢃ0.5 ꢀC.
Acknowledgements
15. (a) Stirchak, E. P.; Summerton, J. E.; Weller, D. D. Nucleic Acids Res. 1989, 17,
6129–6141; (b) Summerton, J. E.; Weller, D. D. WIPO Patent 91/009033, 1991, pp
38–40.
16. Takamizawa, A.; Hirai, K.; Matsui, K. Bull. Chem. Soc. Jpn. 1963, 36, 1214–1220.
17. (a) Gothe, R.; Seyfarth, L.; Schumann, C.; Agricola, I.; Reissmann, S.; Lifferth, A.;
Birr, C.; Filatova, M. P.; Kritsky, A.; Kibirev, V. J. Prakt. Chem. 1999, 341, 369–377;
(b) Schumann, C.; Seyfarth, L.; Greiner, G.; Reissmann, S. J. Pept. Res. 2000, 55,
428–435.
This work was supported by funds from the Spanish Ministerio
´
de Educacion y Ciencia (grants BQU-2001-3693-C02-01 and CTQ-
2004-8275-C02-01) and the Generalitat de Catalunya (2005SGR-
693). S.P.-R. and J.A. acknowledge predoctoral fellowships from the
University of Barcelona (BRD program). The authors also wish to
acknowledge technical support from the Mass Spectrometry and
NMR facilities of the Scientific-Technical Services of the Universitat
de Barcelona.
18. Zapf, C. W.; Goodman, M. J. Org. Chem. 2003, 68, 10092–10097.
19. Mag, M.; Engels, J. W. Nucleic Acids Res. 1989, 17, 5973–5988.
20. Palom, Y.; Grandas, A.; Pedroso, E. Nucleosides Nucleotides 1998, 17, 1177–1182.

S. P´erez-Rentero et al. / Tetrahedron 65 (2009) 1171–1179
1179
21. Lammin, S. G.; Pedgrift, B. L.; Ratcliffe, A. J. Tetrahedron Lett. 1996, 37, 6815–6818.
22. Schneider, S. E.; Bishop, P. A.; Salazar, M. A.; Bishop, O. A.; Anslyn, E. V. Tetra-
hedron 1998, 54, 15063–15086.
27. Guibe´, F. Tetrahedron 1998, 54, 2967–3042.
28. Plum, G. E.; Park, Y.-W.; Singleton, S. F.; Dervan, P. B.; Breslauer, K. J. Proc. Natl.
Acad. Sci. U.S.A. 1990, 87, 9436–9440.
23. Yong, Y. F.; Kowalski, J. A.; Lipton, M. A. J. Org. Chem. 1997, 62, 1540–1542.
24. Alguacil, J.; Pe´rez-Rentero, S.; Robles, J. Unpublished results.
25. Schultz, M.; Hermann, P.; Kunz, H. Synlett 1992, 37–38.
29. Wilson, W. D.; Hopkins, H. P.; Mizan, S.; Hamilton, D. D.; Zon, G. J. Am. Chem.
Soc. 1994, 116, 3607–3608.
30. Durland, R. H.; Kessler, D. J.; Gunnell, S.; Duvic, M.; Pettitt, B. M.; Hogan, M. E.
Biochemistry 1991, 30, 9246–9255.
´
26. Merzouk, A.; Guibe, F. Tetrahedron Lett. 1992, 33, 477–480.