Incorporation of positively charged ribonucleic guanidine linkages into oligodeoxyribonucleotides: Development of potent antisense agents

Article abstract of DOI:10.1016/j.bmcl.2008.02.063

Oligodeoxynucleic acid (21-mer) containing both negatively charged phosphate and positively charged ribonucleic guanidine linkages (RNG/DNA chimera) have been synthesized. DNA binding characteristics and nuclease resistance of RNG/DNA chimeras have been evaluated. Using the bcr-abl oncogene (cause of chronic myeloid leukemia) as a target, the binding of a 21-mer RNG/DNA chimera that includes six RNG's is more than 103.5 stronger than the binding of 21-mer composed solely of DNA.

Full text of DOI:10.1016/j.bmcl.2008.02.063

Available online at www.sciencedirect.com
Bioorganic & Medicinal Chemistry Letters 18 (2008) 2377–2384
Incorporation of positively charged ribonucleic guanidine
linkages into oligodeoxyribonucleotides: Development
of potent antisense agents
Myunji Park, Daniele Canzio and Thomas C. Bruice*
Department of Chemistry and Biochemistry, University of California at Santa Barbara, CA 93106, USA
Received 11 February 2008; revised 22 February 2008; accepted 25 February 2008
Available online 7 March 2008
Abstract—Oligodeoxynucleic acid (21-mer) containing both negatively charged phosphate and positively charged ribonucleic gua-
nidine linkages (RNG/DNA chimera) have been synthesized. DNA binding characteristics and nuclease resistance of RNG/DNA
chimeras have been evaluated. Using the bcr-abl oncogene (cause of chronic myeloid leukemia) as a target, the binding of a 21-mer
RNG/DNA chimera that includes six RNG’s is more than 103.5 stronger than the binding of 21-mer composed solely of DNA.
ꢀ 2008 Elsevier Ltd. All rights reserved.
The possible therapeutic use of oligonucleotide analogs
as effective gene regulatory agents in antisense and anti-
gene approaches has kindled interest in their develop-
ment.1–5 Key goals in the design of such agents include
increased binding affinity while maintaining sequence
specificity, resistance to degradation by cellular nucleas-
es, and improved membrane permeability. Rapid degra-
dation of natural oligonucleotides’ phosphodiester
backbones by cellular nucleases necessitated the creation
of chemically modified oligonucleotides including meth-
ylphosphonate,6,7 methylenemethylimino (MMI),8 and
amide linkages.9 An alternative approach involves
replacing the oligonucleotide backbone entirely, such
as in the case of peptide nucleic acid (PNA),10–12 phos-
phonic ester nucleic acid (PHONA),13 and nucleic acid
analog peptide (NAAP).14 The modified backbones of
these oligonucleotide analogs render them resistant to
nuclease degradation.
the negative phosphodiester linkages of DNA
½–O-ðPO2ꢀÞ-O–ꢁ by positively charged guanidinium link-
ages ½–NH-Cð¼ NH2þÞ-NH–ꢁ.15–17 This replacement al-
ters the electrostatic nature between strands from being
repulsive in native DNA duplexes to attractive in
DNGÆDNA duplexes. These favorable interactions lead
to exceptional binding affinities accompanied by excel-
lent specificities, ideal characteristics for a potential anti-
sense agent.18–21 In addition to these results, the
guanidinium linkage is nuclease resistant,22 and the pos-
itively charged backbones may give rise to increased cell
membrane permeability via electrostatic attraction of
the guanidine moieties to the negative phosphates on
the cell surface.
Because of the potential antisense/antigene aspects of
guanidium linkages of DNG, further studies on the syn-
thesis and properties of guanidium-linked oligonucleo-
tides are warranted.
To increase the free energy of oligonucleotide duplex
formation, the negative charge–charge repulsion in dou-
ble-stranded DNA and RNA can be minimized or elim-
inated if the phosphodiester linkages are replaced by
uncharged or positively charged linkages. Addressing
these design features led us to develop deoxynucleic gua-
nidine (DNG), which is obtained via the replacement of
The interesting differences and similarities between the
nature of DNA and RNA prompted us to explore the
synthesis and properties of ribonucleic guanidine
(RNG), a RNA analog of DNG. Even though RNA
possesses a strong affinity for DNA, its susceptibility
to various nucleases and the labile nature of the phos-
phodiester backbone limit its application.15 RNG
should be better suited as an antisense/antigene agent
because the guanidium linkages are neither suceptible
to cellular nuclease nor chemical degradation under
physiological conditions.
Keywords: Ribonucleic guanidine (RNG); RNG/DNA chimera; Anti-
sense oligonucleotide.
*
Corresponding author. Tel.: +1 805 893 2044; fax: +1 805 893
2229; e-mail: tcbruice@chem.ucsb.edu
0960-894X/$ - see front matter ꢀ 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2008.02.063

2378
M. Park et al. / Bioorg. Med. Chem. Lett. 18 (2008) 2377–2384
Preliminary studies involving binding of cationic penta-
meric ribonucleic guanidine (RNG-U5) to DNA and
RNA have been reported.23,24 Melting temperature
(Tm) and change in circular dichroism (CD) spectra have
been employed to measure the stability of RNG-
U5ÆDNA-A5 and RNG-U5ÆRNA-A5 duplexes. These
data are compared with data for the corresponding
unmodified RNA, DNA duplex to determine the advan-
tage of the stability and structure of the duplexes. The
order of thermal stability observed was RNG-
U5ÆDNA-A5 > RNA-U5ÆRNA-A5 > RNG-U5ÆRNA-A5 >
RNA-U5ÆDNA-A5 > DNA-T5ÆDNA-A5.23,24
RNG/DNA chimeras toward nucleolytic cleavage has
also been investigated.
Phosphoramidite 927 was synthesized (Schemes 1 and 2)
to facilitate the solid-phase synthesis of oligonucleotide
chimeras containing both the standard phosphodiester
and the guanidium linkages.26 Both the required mono-
mers 3 and 4 were prepared from commercially available
uridine. The amino uridine derivative 328 was obtained
by benzylation29 of 20-hydroxy function of 130,31 and
subsequent cleavage of the 50 trityl group of 2. Coupling
of amine 3 with in situ generated carbodiimide from
thiourea monomer 430,31 using HgCl2/Et3N as reported
previously, resulted in Fmoc-protected guanido dimer
5.32 Debenzylation of 5 followed by phosphitylation
produced phosphoramidite 7.33 Finally, Fmoc and silyl
groups of 7 were removed to afford the desired phospho-
ramidite dimer 934 as shown in Scheme 2.
The CD spectra of the RNG-U5ÆRNA-A5 duplex indi-
cate a structure that is different than the reference struc-
tures of the unmodified duplexes.23,24 The CD spectra
and molecular dynamics simulation establish that the
instability of the RNGÆRNA duplex results from the
alteration of the natural conformation of RNA upon
the formation of a duplex with RNG. Molecular
dynamics suggests that the different functional groups
of the RNGÆRNA backbone lead to differences in back-
bone flexibility and length. The duplex compensates for
these differences by adopting high levels of base-pair
propeller, which in turn causes the natural U-O4ꢂ ꢂ ꢂA-
N6 hydrogen bond to break. The propeller leaves the
U-O4 atom in position to hydrogen bond with the A-
N6 atom of the previous base-pair. In contrast, these re-
sults also demonstrate that the sustained B-type DNA
conformation of the RNGÆDNA duplex increases the
stability of this duplex due to the electrostatic attraction.
The affinity of pentameric ribonucleic guanidine (RNG-
U5) to bind to DNA-A5 is greater than the affinity of
DNA-T5.
Phosphoramidite 9 was used as a building block to
introduce guanidium linkages at desired positions in
the chimeric oligonucleotides.35 The chimeras were syn-
thesized using an automated solid-phase synthesizer
with 50-trityl groups on to allow for HPLC purification.
The final detritylated and HPLC purified oligonucleo-
tides were analyzed by mass spectrometry (ESI) and
found to be the desired chimeric products (Table 1).
Thus, RNG/DNA chimeric oligonucleotides, 10–15 hav-
ing either one (10–13) or two (14–15) guanidium link-
ages were successfully prepared. RNG linkages were
introduced at the 30-terminus (11, 13) and middle posi-
tions (10, 12) of chimeras to compare their thermal sta-
bility and resistance to nucleolytic cleavage.
Also, RNG/DNA chimeric oligonucleotides, 19–23
(Table 2), having from three to six guanidium linkages
were successfully obtained by using synthetic synton
1241 in (Scheme 3). Compounds 1041 were prepared by
a previously reported method.30,31 Four–six guanidium
linkages were introduced as blocking 21-mer DNA by
using automated solid-phase DNA synthesizer. The final
oligonucleotides were purified by HPLC and analyzed
by mass spectrometry. Oligonucleotides 19 and 20 have
4 guanidium linkages but in different positions. Also,
oligonucleotides 22 and 23 have 6 guanidium linkages
but in different positions (Table 2). In general, the pos-
itively charged peptide–oligonucleotide conjugates often
show the undesired aggregation formation causing in
difficulties for handling. However, there is no trace of
self aggregation in the RNG/DNA chimeras.22b
The demonstration of a marked preference of RNG for
complementary DNA suggests the synthesis of mixed
RNG and DNA sequences and a study of the binding
of these DNA/RNG chimeras with long complementary
sequences of DNA. Such studies may lead to the synthe-
sis of biologically important antisense/antigene agents.
Ider developed oligonucleotides to target the bcr-abl
oncogene, which causes chronic myeloid leukemia
(CML) and bcr-abl-positive acute lymphoblastic leuke-
mia (ALL).25 It was demonstrated that anti-bcr-abl oli-
gonucleotides specifically inhibit bcr-abl mRNA
expression up to 70% in hematopoietic cell line and pri-
mary CML. Substitution of uridine–uridine phosphodi-
ester linkages with uridine–uridine guanidium linkages
on the oligonucleotide strands to target the bcr-abl
oncogene (DNA/RNG chimeras) would be expected to
be the more potent candidate of antisense against bcr-
abl oncogene mRNA.
All chimeric oligonucleotides were hybridized with their
respective complementary strands in a 1:1 mol ratio in
phosphate buffer (10 mM Na2HPO4, pH 7.1, 0/10/
100 mM NaCl). UV-thermal denaturation studies were
performed to observe the effect of the guanidium linkage
on the formation of duplexes.
In this paper, we report the synthesis of 21 base-
paired RNG/DNA chimeras, having mixed anionic
phosphodiester linkages of DNA and cationic guanid-
inium linkages of RNG based on the oligonucleotide
sequence to target bcr-abl oncogene. The hybridiza-
tion properties of the RNG/DNA chimeras with com-
plementary DNA have been evaluated using
spectroscopic techniques. The stability of these
All of the temperature versus absorbance curves were
sigmoidal (Fig. 1) showing that the double helix forma-
tion is cooperative. For the duplex formed with RNG/
DNA chimeras (1–6) and complementary DNA, no
apparent difference in Tm occurs between the control

M. Park et al. / Bioorg. Med. Chem. Lett. 18 (2008) 2377–2384
2379
Scheme 1. Preparation of dimer synthon.
Scheme 2. Preparation of dimer synthon 9.
DNA–DNA and RNG/DNA–DNA at all salt concen-
trations. Furthermore, it is observed that the incorpo-
ration of one or two guanidium linkages has no
significant effect on the hybridization properties with
complementary DNA sequences at all salt concentra-
tions (Table 3).

2380
M. Park et al. / Bioorg. Med. Chem. Lett. 18 (2008) 2377–2384
Table 1. Sequences used for RNG/DNA characterization
Compound
RNG/DNA chimeras
Yields (%)
Calculateda (M)
Foundedb (m/z)
13
14
15
16
17
18
50-AAGGGCUgUTTGAACTCTGCTT-30
50-AAGGGCTTTTGAACTCTGCUgU-30
50-AGGGCUgUTTGAACTCTGCTTT-30
50-AGGGCTTTTGAACTCTGCTUgU-30
50-AAGGGCUgUUgUGAACTCTGCTT-30
50-AGGGCUgUUgUGAACTCTGCTTT-30
42
40
45
43
31
29
6588.33
6588.33
6565.45
6565.45
6533.40
6510.52
6589.54
6589.61
6566.70
6566.61
6534.50
6511.74
Oligonucleotides RNG/DNA chimera with one or two guanidium linkages guanidinium linkages indicated as ‘g’.
a Calculated molecular weight (M) of oligonucleotides.
b Determined molecular weight (M+H)+ of oligonucleotides by mass spectrometry spectra.
Table 2. Oligonucleotides RNG/DNA chimera with four–six guanidium linkages guanidinium linkages indicated as ‘g’
Compound
RNG/DNA chimeras
Yields (%)
Calculateda (M)
Foundedb (m/z)
19
20
21
22
23
24
50-AGGGTUgUgUgUgUTAACTCTGCTT-30
50-AGGGTUgUgUgUTTAACTCTGCUgU-30
50-AGGGUgUgUgUgUgUTAACTCTGCTT-30
50-AGGGUgUgUgUgUgUgUAACTCTGCTT-30
50-AGGGUgUgUgUgUgUTAACTCTGCUgU-30
50-AGGGTTTTTTTAACTCTGCTT-30
28
26
22
19
21
6311.18
6313.16
6292.24
6273.33
6275.31
Control
6311.98
6313.61
6292.91
6273.98
6275.76
a Calculated molecular weight (M) of oligonucleotides.
b Determined molecular weight (M+H)+ of oligonucleotides by mass spectrometry spectra.
Scheme 3. Preparation of synthons 12.
The stability of the RNG/DNA–DNA duplexes increases
with the increase in salt concentration (0–100 mM NaCl),
as is seen in DNA–DNA duplexes. This indicates that the
incorporation of one or two guanidium linkages in DNA
does not affect the overall electrostatics of duplex forma-
tion with complementary DNA strand.

M. Park et al. / Bioorg. Med. Chem. Lett. 18 (2008) 2377–2384
2381
oligonucleotides 1–6 were subjected to nucleolytic cleav-
age by exonuclease I and the hydrolyzate was analyzed
by RP-HPLC.38 Natural unmodified oligonucleotides
of same sequences were used as controls.
0.14
0.12
0.1
0.08
0.06
0.04
0.02
0
Under the conditions employed, the control oligonucleo-
tides were readily hydrolyzed to shorter length products
within 1 h of incubation; however, the RNG/DNA chi-
meras 2 and 5 were unaltered even after 12 h of incuba-
tion. The RNG/DNA chimeras 1, 3, 4, 6 (rt; retention
time = 24.3, 25.0, 24.9, 25.6), which contain one or two
guanidinium linkages at the center of the oligonucleo-
tide, was partially hydrolyzed after 1 h (rt = 21.3 and
24.3, 20.3 and 25.0, 22.2 and 24.9, 23.2 and 25.6) of incu-
bation and remained further unaltered even after 12 h.
These observations clearly indicate that the RNG/
DNA oligonucleotides with positively charged guanidi-
um linkages at 30-terminus, 2 and 5, are totally stable
to cleavage by nucleolytic enzyme exonuclease I.
20
40
60
80
Wavelength
DNA/DNA
DNA/16
DNA/13
DNA/17
DNA/14
DNA/18
DNA/15
Figure 1. Melting studies 1; thermal denaturation curves for RNG/
DNA chimeras and complementary DNA duplexes. Absorbance was
measured at 260 nm; the concentration of each strand was 6 lM in
10 mM Na2HPO4, 100 mM NaCl, pH 7.1.
The melting temperatures and thermodynamic parame-
ters of chimeras (19–23) are provided in Figure 2 and
Table 3. All experimental conditions employed in melt-
ing studies of DNA/RNG chimeras 19–23 were exactly
the same as those of melting studies of DNA/RNG chi-
meras with one or two guanidium linkages.
The DG0 values (at 25 ꢁC) calculated using the van’t
Hoff enthalpy values for transitions36,37 involving
RNG/DNA chimera–DNA and DNA–DNA duplexes,
showed that the free energy for the formation of
RNG/DNA–DNA duplex is about the same as for
DNA–DNA duplex. This supports the idea that little
structural difference exists between DNA containing
one or two guanidium linkages and DNA composed en-
tirely of phosphodiester linkages.
As expected, chimeras 19 and 20 with four charged guani-
dium linkages in 21-mer DNA binds to complementary
DNA with much higher affinity (Tm = 65.0, 65.1 ꢁC) than
DNA (48 ꢁC) (Table 3). Also, chimeras 21–23, where five
or six phosphodiester linkages have been replaced by gua-
nidium linkages in 21-mer DNA/RNG, bind to comple-
mentary DNA with unprecedented higher affinity
(Tm = 78.7, 82.4, 83.0 ꢁC). As shown in Table 2, chimeras
19 and 20, or chimeras 22 and 23 have the same number of
guanidium linkages but in different positions. The 30-end
phosphodiester linkage has been substituted with guani-
dium linkage in chimeras 20 and 23.
Exonuclease I digests single-stranded DNA by catalyz-
ing the hydrolysis of the phosphodiester linkages from
the 30 to 50-terminus.39,40 Thus, it is assumed that upon
modification of the phosphodiester linkage at the 30-ter-
minus, the oligonucleotide could be resistant to
exonuclease digestion. To investigate this, RNG/DNA
Table 3. Melting temperatures and thermodynamic parameters for helix–coil transitions of RNG/DNA chimeras with complementary DNA
RNG/DNA chimera
Duplexa
0 mM NaCl
10 mM NaCl
100 mM NaCl
b
b
b
Tm (ꢁC)
ꢀDG025 kJ=mol
Tm (ꢁC)
ꢀDG025 kJ=mol
Tm (ꢁC)
ꢀDG025 kJ=mol
13
14
15
16
17
18
25
26
33.1
32.9
31.8
32.6
33.3
33.1
32.7
35.9
35.6
34.5
35.3
36.1
35.9
35.4
38.3
37.7
37.3
38.5
38.0
38.6
37.0
41.5
40.8
40.4
41.7
41.2
41.8
40.1
49.0
49.5
48.6
50.1
51.3
49.6
48.0
53.1
53.6
52.7
54.3
55.6
53.7
52.0
27
28
29
30
Control
100 mM NaCl
b
Tm (ꢁC)
ꢀDG025 kJ=mol
19
20
21
22
23
31
32
33
34
35
65.0
65.1
78.5
82.4
83.0
46
60.3
60,3
63.0
72.1
72.2
51.6
Control
Thermodynamic parameters were calculated by the method of Gralla and Crothers.36,37
a Absorbance was measured at 260 nm in phosphate buffer; the concentration of each strand was 6 lM.
b The reported Tm values are an average of three experiments ( 0.2).38

2382
M. Park et al. / Bioorg. Med. Chem. Lett. 18 (2008) 2377–2384
solid-phase synthesis techniques. Substitution of one
0.2
0.15
0.1
or two positively charged RNG into a 21 DNA se-
quence provides chimeras that bind to a complemen-
tary DNA strand with similar (or a little higher)
stability is found for the like DNA/DNA binding.
This is a true at various ionic strengths. Incorpora-
tion of positively charged guanidium linkages at the
30-terminus of the DNA prevents digestion by exonu-
clease I.
0.05
0
The results of melting temperatures of RNG/DNA chi-
meras in which four–six phosphodiester linkages are re-
placed by guanidium linkages, show that four
guanidium linkages in 21-mer DNA are long enough
to enforce the affinity with complementary DNA. Fur-
thermore, oligonucleotide chimeras which have five or
six guanidium linkage in 21-mer have tremendously
higher Tm temperature compared to natural DNA du-
plex. It has been strongly suggested that thermodynam-
ically favorable DNA/RNG chimeras may serve as
potent antisense candidates. In vitro experiments to
evaluate the antisense effect of RNG/DNA chimeras will
follow.
20
40
60
80
Wavelength
DNA/DNA
DNA/21
DNA/19
DNA/22
DNA/20
DNA/23
Figure 2. Melting studies 2; thermal denaturation curves for RNG/
DNA chimeras and complementary DNA duplexes. Absorbance was
measured at 260 nm; the concentration of each strand was 6 lM in
10 mM Na2HPO4, 100 mM NaCl, pH 7.1.
Thermodynamic calculations were performed in order
to describe the duplex properties more quantitatively.
Standard free energies (DG0) for duplex formation are
presented in the following equations:
Acknowledgment
0
This work was supported by a Grant from the National
Institute of Health (DK09171).
DNAðcontrolÞþDNA
ꢀ
DG ¼ꢀ51:6 kJ=mol DNAꢂDNA
ð1Þ
DG0¼ꢀ60:3 kJ=mol
RNG=DNAchimera; 20þDNA
ꢀ
20ꢂDNA
ð2Þ
References and notes
DG0¼ꢀ63:0 kJ=mol
RNG=DNAchimera; 21þDNA
RNG=DNAchimera; 23þDNA
ꢀ
21ꢂDNA
ð3Þ
1. Uhlmann, E.; Peyman, A. Chem. Rev. 1990, 90, 543.
2. De Mesmaeker, A.; Haener, R.; Martin, P.; Moser, H. E.
Acc. Chem. Res. 1995, 28, 366.
3. Agrawal, S.; Zhao, Q. Y. Curr. Opin. Chem. Biol. 1998, 2,
519.
DG0¼ꢀ72:2 kJ=mol
ꢀ
23ꢂDNA
ð4Þ
4. Thoung, N. T.; Helene, C. Angew. Chem., Int. Ed. Engl.
1993, 32, 666.
5. Cook, P. D. Nucleosides Nucleotides 1990, 18, 1141.
6. Stein, C. A.; Cheng, Y.-C. Science 1993, 261, 1004.
7. Tseng, B. Y.; Ts’o, P. O. P. Antisense Res. Dev. 1995, 5,
251.
8. Morvan, F.; Sanghvi, Y. S.; Perbost, M.; Vasseur, J.-J.;
Bellon, L. J. Am. Chem. Soc. 1996, 118, 255.
9. De Mesmaeker, A.; Lesueur, C.; Bevierre, M.-O.; Wald-
ner, A.; Fritsch, V.; Wolf, R. M. Angew. Chem., Int. Ed.
Engl. 1996, 35, 2790.
The DDG0 of ꢀ11.4 kJ/mol (Eq. 3 DGꢁ–Eq. 1 DG0)
and ꢀ20.6 kJ/mol (Eq. 4 DG0–Eq. 1 DG0) translates
into an increase in binding over multi orders of mag-
nitude. Results clearly demonstrate the over 103.5 tigh-
ter binding of RNG/DNA chimeras 22, 23 to its DNA
templates compared to natural DNA duplex. The tre-
mendous increase in the free energy of binding of
DNA/RNG chimeras 19–23 (Table 3) support the
powerful attraction between the positively charged
RNG guanidium groups and the negatively charged
phosphates of DNA. This increase is attributed to
both the formation of hydrogen bonds37b in combina-
tion with electrostatic attraction of RNG for DNA.
This suggests that strategically substituted guanidi-
um-linked uridines may yield powerful antisense
oligonucleotide.
10. Bohler, C.; Nielsen, P. E.; Orgel, L. E. Nature 1995, 376,
578.
11. Veselkov, A. G.; Demidov, V. V.; Frank-Kamenetiskii, N.
D.; Nielsen, P. E. Nature 1996, 379, 214.
12. Uhimann, E.; Peyman, A.; Breipohl, G.; Will, D. W.
Angew. Chem., Int. Ed. Engl. 1998, 37, 2796.
13. Peyman, A.; Uhlmann, E.; Wagner, K.; Augustin,
S.; Breipohl, G.; Will, D. W.; Schafer, A.; Wall-
meier, H. Angew. Chem., Int. Ed. Engl. 1996, 35,
2636.
14. Fujii, M.; Yoshida, K.; Hidaka, J. Bioorg. Med. Chem.
Lett. 1997, 7, 637.
15. Dempcy, R. O.; Browne, K. A.; Bruice, T. C. J. Am.
Chem. Soc. 1995, 117, 6140.
16. Dempcy, R. O.; Browne, K. A.; Bruice, T. C. Proc. Natl.
Acad. Sci. U.S.A. 1995, 92, 6097.
A method for the incorporation of a positively
charged ribointernucleoside into negatively charged
DNA has been demonstrated. The insertion of the
guanidium linkage was accomplished using the stan-
dard DNA phosphoramidite chemistry and automated

M. Park et al. / Bioorg. Med. Chem. Lett. 18 (2008) 2377–2384
2383
17. Dempcy, R. O.; Almarsson, O.; Bruice, T. C. Proc. Natl.
Acad. Sci. U.S.A. 1994, 91, 7864.
113 mg, 71%). m/z (FAB) 1325.81 calculated for
C49H61N9O12P (M+H)+ 1325.67.
18. Browne, K. A.; Dempcy, R. O.; Bruice, T. C. Proc. Natl.
Acad. Sci. U.S.A. 1995, 92, 7051.
19. Linkletter, B. A.; Szabo, I. E.; Bruice, T. C. J. Am. Chem.
Soc. 1999, 121, 388.
20. Linkletter, B. A.; Szabo, I. E.; Bruice, T. C. Bioorg. Med.
Chem. 2000, 8, 1893.
21. Linkletter, B. A.; Szabo, I. E.; Bruice, T. C. Nucleic Acids
Res. 2001, 29, 2370.
22. (a) Barakar, D. A.; Bruice, T. C. Proc. Natl. Acad. Sci.
U.S.A. 1998, 95, 11047; (b) Barakar, D. A.; Kwok, Y.;
Bruice, T. W.; Bruice, T. C. J. Am. Chem. Soc. 2000, 122,
5244.
23. Park, M.; Bruice, T. C. Bioorg. Med. Chem. Lett. 2005, 15,
3247.
24. Park, M.; Toporowski, J. W.; Bruice, T. C. Bioorg. Med.
Chem. 2006, 14, 1743.
25. Dwaine, A.; Susan, J.; Yinghui, L.; Kiran, K.; Khalil, A.;
Michael, A.; David, C. Biochemistry 2003, 42, 7967.
26. Arya, D. P.; Bruice, T. C. J. Am. Chem. Soc. 1998, 120,
6619.
27. Compound 9: Compound 8 (294 mg, 0.24 mmol) was
dissolved in 1.0 M TBAF in THF (0.5 ml, 0.5 mmol) and
stirred at room temperature for 16 h. Acetic acid (0.3 ml)
was added, diluted with H2O, and then Et2O was added.
The aqueous phase was concentrated in vacuo. The
residue was purified by RP-HPLC (Altech Macrosphere
RP C8 column 0–60% acetonitrile in 3% aqueous acetic
acid) (195 mg, 81%). m/z (FAB) 999.0 calculated for
C49H61N9O12P (M+H)+ 999.03.
28. Compound 3: Compound 2 (60 mg, 0.1 mmol) was dis-
solved in 80% aqueous acetic acid and stirred at room
temperature for 24 h. The solvent was removed in vacuo.
Extraction with ethyl acetate followed by dilution with
water, brine, drying over anhydrous Na2SO4 and evapo-
rating under reduced pressure. The crude residue was
subjected to silica gel column chromatography using
dichloromethane/methanol (50:1–1:2) solvent system to
give the pure product as a brittle white form (32 mg, 80%).
m/z (FAB) 448.71 calculated for C49H61N9O12P (M+H)+
448.63.
29. Wilson, T. M., Kocrenski, P., Jarowicki, K., Issac, K.,
Hitchcock, P. M., Faller, A., Campbell, S. F. Tetrahe-
dron 1990, 46, 1767. Compound 2: Compound 1 (1.9 g,
30 mmol) solution in TFH was added dropwise to a
stirred suspension of NaH (142 mg, 30.2 mmol) in
TFH. After 10 min, benzyl bromide (664 mg,
3.83 mmol) and KI (38 mg) in TFH were added slowly.
The suspension was stirred at 20 ꢁC for 1 h and then
poured into water. Extraction with ethyl acetate
followed by the addition of water, brine, drying over
anhydrous Na2SO4 and evaporated under reduced
pressure. The crude residue was subjected to silica gel
column chromatography using AcOEt/hexane (1:5–1:1)
solvent system to give the pure product as a brittle
white form (1.4 g, 65%). m/z (FAB) 721.71 calculated
for C49H61N9O12P (M+H)+ 720.94.
33. Compound 6: A mixture of 5 (60 mg, 0.05 mmol) and 10%
Pd/C (10 mg) in EtOH was stirred under 50 psi by parr
hydrogenation for 4 h. The catalyst was removed through
Celite pad and the filtrate was concentrated in vacuo. The
residue was purified by silica gel column chromatography
using dichloromethane/methanol (5:1–1:2) solvent system
to give the pure product as a brittle white form (56 mg,
91%). m/z (FAB) 1235.2 calculated for C66H80N7O13Si2
(M+H)+ 1235.55.
Compound 7: To a mixture of compound 6 (124 mg,
0.1 mmol) and DIEA (0.15 mmol) in dried dichlorometh-
ane was added [chloro-(diisopropylamine)-b-cyanoethoxy-
phosphine] (0.12 mmol) and stirred for 2 h. The solvent
was removed in vacuo. Extraction with ethyl acetate
followed by dilution with water, brine, drying over
anhydrous Na2SO4 and evaporating under reduced
pressure. The crude residue was subjected to column
chromatography using AcOEt/hexane (1:5–1:2) solvent
system to give the pure product as a bright yellow form
(112 mg, 77%). m/z (FAB) 1449.6 calculated for
C76H99N7O14PSi2 (M+H)+ 1449.80.
34. Compound 8: Compound 7 (145 mg, 0.1 mmol) was
dissolved in 28% NH4OH/EtOH (3:1) and kept at room
temperature for 12 h. The mixture was filtered through
the glass filter. The filtrate was diluted with ethyl acetate,
which was diluted with water, brine, dried over anhy-
drous Na2SO4 and concentrated in vacuo. The crude
residue was purified by silica gel column chromatogra-
phy (AcOEt/hexane 1:2–1:1) solvent system to give the
pure product as a bright yellow form (99 mg, 81%). m/z
(FAB) 1227.2 calculated for C61H89N9O12PSi2 (M+H)+
1227.56.
35. Oligonucleotide synthesis and purification: All modified
oligonucleotides were synthesized on 1.3 lmol scale on
Pharmacia Gene Assembler Plus DNA synthesizer. Stan-
dard DNA synthesis condition was employed, viz., CPG
support and base protected 50-O-(4,40-dimethoxytri-
tyl)deoxyribonucleoside-30-[O-(diisopropylamino)-b-cyano-
ethylphosramidite] monomers. The phosphoramidite
activated oligomer 9.12 was used at an extra monomer
position, in order to introduce a guanidium linkage into
Oligonucleotides. The standard synthesis cycle was mod-
ified to perform an extended coupling (15 min) during the
coupling the modified phosphoramidite oligomer 9.12; a
coupling efficiency of >95% was observed for this step.
36. (a) Marky, L. A.; Breslauer, K. J. Biopolymers 1987, 26,
2601; (b) Breslauer, K. J. In Methods in Molecular biology;
Agrawal, S., Ed.; Humana: Totowa, 1995; Vol. 26, p 347.
37. (a) Gralla, J.; Crothers, D. M. J. Mol. Biol. 1973, 78, 301;
(b) Szabo, I. E.; Bruice, T. C. Bioorg. Med. Chem. 2004,
12, 4233.
38. Johnson, W. T.; Zhang, P.; Bergstrom, D. E. Nucleic Acids
Res. 1997, 25, 559.
39. Exonuclease I digestion studies: Oligonucleotides 1–6,
with phosphodiester or guanidium linkages were treated
with exonuclease I (USB). A typical reaction mixture
contained the following in 200 ll: 67 mM Tris (pH 8.5),
6.7 mM MgCl2. 20 mM 2-mercaptoethanol, ꢃ0.2 OD of
oligonucleotide, and ꢃ20 U of exonuclease I. The
reactions were incubated at 37 ꢁC. Aliquots (40 ll) were
taken at different time intervals (0.1, 3, 6 and 12 h),
quenched by rapid freezing in dry ice-2-propanol bath,
and stored frozen until HPLC analysis. Reaction mix-
tures were analyzed on C-8 RP-HPLC column (Altech,
7 l, 4.6· 250 mm) using a gradient of 0.5%/mim of
CH3CN in 0.1 M TEAA, pH 7.0, for 50 min at a flow
rate of 1 mL/min. Reactions without enzyme were run
30. Kojima, N.; Bruice, T. C. Org. Lett. 2000, 2, 81.
31. Kojima, N.; Szabo, I. E.; Bruice, T. C. Tetrahedron 2002,
58, 867.
32. Compound 5: To a mixture of compound 3 (50 mg,
0.12 mmol) and 4 (102 mg, 0.12 mmol) in DMF were
added DIEA (0.3 mmol) and HgCl2 (41 mg, 0.15 mmol)
for 3 h. The mixture was filtered through a Celite pad. The
filtrate was diluted with ethyl acetate, which was diluted
with water, brine, dried over anhydrous Na2SO4 and
concentrated in vacuo. The crude residue was purified by
silica gel column chromatography (AcOEt/hexane 1:2,

2384
M. Park et al. / Bioorg. Med. Chem. Lett. 18 (2008) 2377–2384
for each oligonucleotide and analyzed by HPLC as
controls.
40. Hemavathi, C.; Bruice, T. C. Bioorg. Med. Chem. Lett.
2001, 11, 2423.
41. To a mixture of oligonucleotides 10 (0.1 mmol) and
DIEA (0.15 mmol) in dried dichloromethane was
over anhydrous Na2SO4 and evaporating under
reduced pressure. The crude oligonucleotide was
dissolved in 1.0 M TBAF in THF (0.3 mmol) and
stirred at room temperature for 16 h. Acetic acid
(0.3 ml) was added, diluted with H2O, and Et2O was
added. The aqueous phase was concentrated in vacuo.
The residue was purified by RP-HPLC (Altech
Macrosphere RP C8 column 0–40% acetonitrile in
3% aqueous acetic acid). Mass spectrometry data for
compound 10, 11, 12.
added
[chloro-(diisopropylamine)-b-cyanoethoxyphos-
phine] (0.12 mmol) and stirred for 2 h. The reaction
solvent was removed in vacuo. Extraction with ethyl
acetate followed by dilution with water, brine, drying
n
3
4
5
6
Compound m/z
Calculated for
(M+H)+
m/z
(FAB)
Calculated for
(M+H)+
m/z
(FAB)
Calculated for
(M+H)+
m/z
(FAB)
Calculated for
(M+H)+
(FAB)
10
11
12
C89H146N17O19PSi4
C107H181N22O23PS5
C125H216N27O27PS6
C143H251N32O31PS7
1871.01 1871.51
2283.25 2283.26
2696.73 2696.50
3111.75 3111.32
C98H163N19O20PSi4
C116H198N24O24PS5
C134H233N29O28PS6
C152H268N34O32PS7
3311.8
C110H170N34O322P
2.71.01 2070.11
C74H107N19O20
1613.77 1613.73
2484.37 2484.36
C86H128N24O24
1911.93 1911.92
2897.95 2898.61
C98H149N29O28
2212.09 2212.38
3311.5
P
P
P
2511.75 2511.25