Modified α-β chimeric oligoDNA bearing a multi-conjugate of 2,2-bis(hydroxymethyl)propionic acid-anthraquinone-polyamine exhibited improved and stereo-nonspecific triplex-forming ability

Article abstract of DOI:10.1039/b514325j

Novel α-β chimeric oligonucleotides bearing a propionic acid derivative of an anthraquinone-polyamine conjugate in the "linker" region sequence-specifically formed a substantially stable alternate-stranded triplex with dsDNA almost regardless of the stere

Full text of DOI:10.1039/b514325j

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Modified a–b chimeric oligoDNA bearing a multi-conjugate of
2,2-bis(hydroxymethyl)propionic acid–anthraquinone–polyamine
exhibited improved and stereo-nonspecific triplex-forming ability{
A. T. M. Zafrul Azam, Tomohisa Moriguchi and Kazuo Shinozuka*
Received (in Cambridge, UK) 10th October 2005, Accepted 9th November 2005
First published as an Advance Article on the web 23rd November 2005
DOI: 10.1039/b514325j
In our continuing studies to bring about further improvement in
the affinity of chimera TFO to its target sequence, we have
designed a new type of modified a–b chimera TFO, GK-354 and
GK-358. In these TFOs, a normal nucleoside unit in the ‘‘linker’’
region is substituted with a multi-conjugate of a non-nucleosidic
unit, 2,2-bis(hydroxymethyl)propionic acid, and an intercalator–
polyamine complex (Fig. 1). Although the TFOs become a set of
diastereomers due to the presence of a stereogenic carbon atom in
the non-nucleosidic unit, a UV-melting experiment revealed that
the TFOs have almost equal and remarkable triplex-stabilizing
ability. Here, we would like to report the synthesis of the
multi-conjugate of 2,2-bis(hydroxymethyl)propionic acid–anthra-
quinone–polyamine and its incorporation into the chimera TFO,
as well as the thermal stability of the alternate-stranded triplex
formed with dsDNA and the modified chimera TFO.
Novel a–b chimeric oligonucleotides bearing a propionic acid
derivative of an anthraquinone–polyamine conjugate in the
‘‘linker’’ region sequence-specifically formed a substantially
stable alternate-stranded triplex with dsDNA almost regardless
of the stereochemistry of the derivative.
Triplex-forming oligonucleotide (TFO) has been attracting great
interest as a possible new therapeutic agent since it would regulate
gene expression through direct interaction with genomic dsDNA.
Indeed, oligonucleotides forming a triplex with certain gene
promoters can modulate the level of transcription of that gene
(anti-gene strategy).1 One major drawback of the strategy is that
the thermal stability of the resulting triplex under physiological
conditions is not very high compared to that of the duplex having
an analogous sequence. Furthermore, sufficiently long polypurine
tracts in one strand of genomic DNA, which serves as the target
for TFO, are not always found in a biologically important
sequence. Considerable effort has been devoted to overcoming
these limitations of the anti-gene strategy. Accordingly, a unique
type of TFO consisting of 39-39 conjugated polythymidylate was
reported by Horne and Dervan.2 The TFO forms a so-called
alternate-stranded triplex with an adjacent and alternating
homopurine strand in the manner of parallel orientation. As a
result, the TFO could expand its target and form a sufficiently long
triplex with improved thermal stability. Since then, a number of
39-39 or 59-59 conjugated oligopyrimidylates have been synthesized
having higher affinity to the target dsDNA through forming an
alternate-stranded triple helix.3
The synthetic procedure of the key compounds, bi-functional
5-tris(aminoethyl)anthraquinone phosphoramidite reagents (7) is
Meanwhile, we have reported the synthesis and alternate-
stranded triplex formation of a chimeric oligoDNA composed of a
contiguous b-anomeric polypyrimidine strand and an a-anomeric
polypyrimidine strand.4 Since the thermal stability of the resulting
triplex was just above the physiological temperature even at pH
6.1, we subsequently developed several modified a–b chimera
TFOs possessing either a novel dinucleoside phosphotriester unit
conjugated with an anthraquinone moiety at the internucleotidic
linkage in a stereospecific manner5 or an anthraquinone-attached
nucleobase, both in the linker region of the chimera DNA.6
Fig. 1 The sequence and the structure of a–b chimeric TFOs (GK-300,
GK-354 and GK-358). In the sequence, italic letters represent the
a-anomeric polypyrimidylate component and roman letters represent the
b-anomeric polypyrimidylate component, respectively. The linker portion
of the chimera TFO consists of a b-anomeric 59-AACC-39 sequence in
GK-300 whereas AC of the portion is replaced with a non-nucleotidic
anthraquinone bearing unit in GK-354 and GK-358. Duplex containing
ODN-2 and ODN-3 represent a full-matched target whereas ODN-4 and
ODN-5 represent a single mismatched target of TFO.
Department of Chemistry, Faculty of Engineering, Gunma University,
1-5-1, Tenjin-cho, Kiryu, Gunma, 376-8515, Japan.
E-mail: sinozuka@chem.gunma-u.ac.jp; Fax: 81-277-30-1321;
Tel: 81-277-30-1320
{ Electronic supplementary information (ESI) available: UV-melting
curves of triplex at pH 6.5 and 7.0 (Figs. S1–S4). See DOI: 10.1039/
b514325j
This journal is ß The Royal Society of Chemistry 2006
Chem. Commun., 2006, 335–337 | 335

View Article Online
summarized in Scheme 1. In brief, 5-chloroanthraquinolyl diamine
derivative (2), which was readily prepared from 1,5-dichloroan-
thraquinone (1) and n-alkyl diamine, was condensed with
2,2-bis(hydroxymethyl)propionic acid by the action of dicyclo-
hexylcarbodiimide (DCC) and 1-hydroxybenzotriazole in dry
pyridine.7 The obtained conjugate (3) was reacted under reflux
with tris(2-aminoethyl)amine in toluene to give compound (4). The
trifluoroacetyl-protected compound (5) was prepared by the
treatment of (4) with trifluoroacetic acid ethyl ester in the presence
of triethylamine in dry dichloromethane. The obtained conjugate
(5) was reacted with dimethoxytrityl chloride to achieve the
monotritylated desired compound (6) as the major product along
with a small amount of the ditritylated by-product. Compound (6)
was further converted to the corresponding phosphoramidite
derivative{ (7) by a standard procedure8 for the automated solid-
phase DNA synthesis.
Meanwhile, unnatural a-29-deoxythymidine was prepared from
natural b-29-deoxythymidine by an epimerization reaction.9
a-29-Deoxy-5-methylcytidine was prepared from a-4-triazolyl-
29-deoxythymidine as described previously.10 The nucleosides were
converted to the corresponding phosphoramidite derivatives in the
same manner as above.
The incorporation of the phosphoramidite derivative bearing
conjugated anthraquinone (7) into the ‘‘linker’’ region of the
chimera TFO was accomplished with an automated DNA
synthesizer (ABI-392) on a 0.2 mmol scale starting from CPG-
bound a-deoxynucleoside which was prepared according to the
reported procedure.11 The coupling reaction of the phosphorami-
dite (0.3 M) was carried out for 360 s with a double coupling
procedure. The coupling yield estimated from the conventional
trityl assay was about 95% for both GK-354 and GK-358. The
incorporation of other unnatural nucleoside units into the chimera
TFO was also accomplished with an extended coupling period
(360 s). The obtained CPG-bound oligomer was treated with
concentrated ammonia at 55 uC for 12 h. Then the oligomer was
purified by reversed-phase HPLC.
Fig. 2 shows the HPLC profile of GK-354. In the DMTr-ON
form (Fig. 2a), GK-354 gave a single product peak at retention
time 15.7 min. After isolation and detritylation, the sample was
again analyzed by HPLC. This time, the sample gave two peaks at
retention time 30.9 min and 33.0 min with almost equal intensity.
Because the introduced anthraquinone derivative has a stereogenic
carbon atom, these could be diastereomers of the desired chimera
DNA. Therefore, we isolated these two peaks separately and
designated the faster eluate as GK-354A and the slower eluate as
GK-354B. The GK-358A and GK-358B were isolated similarly and
the oligomers were further purified by ethanol precipitation and
Sephadex G-25 gel filtration.§
The formation and thermal stability of the complex formed with
the isolated modified TFO and the dsDNA consisting of ODN-2
and ODN-3 were examined by UV-melting experiments under
near physiological conditions (pH 6.5 and 7.0). In the experiments,
all complexes exhibited a two-phase transition upon increasing the
temperature (Figs. S1–S412). The lower transitions of all triplexes
formed with the modified TFO, however, shifted towards a higher
Scheme 1 Preparation of 5-tris(aminoethyl)anthraquinone bearing phos-
phoramidite reagents for the automated solid-phase oligonucleotide
synthesis. a) 3 equiv. of NH2-(CH2)n-NH2, reflux in xylene, 30 min
(56.5%, n = 2; 56.3%, n = 6); b) 1.3 equiv. of 2,2-bis(hydroxymethyl)pro-
pionic acid, 2.0 equiv. of DCC and 1.5 equiv. of HOBt in dry pyridine, r.t.,
4 h (60.4%, n = 2; 81.3%, n = 6); c) 4.0 equiv. of tris(2-aminoethyl)amine,
reflux in toluene, 6 h (51.6%, n = 2; 65.0%, n = 6); d) 5.0 equiv. of
CF3COOC2H5, 10 equiv. of TEA in dry CH2Cl2, r.t., 3 h (60.7%, n = 2;
85.4%, n = 6); e) 1.2 equiv. of DMTr-Cl, 1.5 equiv. of DIPEA and
0.05 equiv. of DMAP in dry pyridine, r.t., 3 h (66.1%, n = 2; 74.0%, n = 6);
f) 2.5 equiv. of (iPr2N)(NCCH2CH2O)-PCl, 3.0 equiv. of DIPEA in dry
CH2Cl2, r.t., 30 min (70.1%, n = 2; 81.0%, n = 6).
Fig. 2 The HPLC profile of GK-354. (a) DMTr-ON form; conditions: A
buffer = 100 mM TEAA and B buffer = 100% acetonitrile; gradient: time
in min (B buffer%), 0 (20), 10 (30), 30 (30), 35 (20). (b) DMTr-OFF form;
conditions: A buffer = 50 mM TEAA and B buffer = 70% acetonitrile in
50 mM TEAA; gradient: time in min (B buffer%), 0 (18), 35 (25), 45 (25),
50 (18). For both cases the flow rate was 1 ml/min and the column was
Wakosil 5C18 (ø 4.6 mm 6 250 mm).
336 | Chem. Commun., 2006, 335–337
This journal is ß The Royal Society of Chemistry 2006

View Article Online
Table 1 Melting temperature (Tm) of the triplexes and the duplex at pH 6.5
a
Tm (Full-match)
a
Tm (Mismatch)
Triplexd
Duplexd
DTm
Triplexd
Duplexd
DTm
b,d
c,d
TFO
GK-300
GK-354A
GK-354B
GK-358A
GK-358B
a
35.8 (23.6)
50.3 (40.4)
51.2 (40.2)
46.4 (35.5)
46.6 (36.4)
72.2 (71.5)
71.8 (71.5)
71.7 (71.3)
71.7 (71.2)
71.6 (71.2)
—
26.2 (16.3)
42.4 (33.3)
42.2 (33.6)
39.1 (28.8)
39.4 (30.1)
72.2 (71.7)
72.1 (71.5)
71.9 (71.4)
71.9 (71.4)
71.8 (71.5)
29.6 (27.3)
27.9 (27.1)
29.0 (26.6)
27.3 (26.7)
27.2 (26.3)
b
14.5 (16.8)
15.4 (16.6)
10.6 (11.9)
10.8 (12.8)
Tm values (uC) were determined by computer fitting of the first derivative of the absorbance with respect to 1/T. DTm indicates the
c
deviation from the Tm value of the corresponding triplex consisting of GK-300 and the duplex. DTm indicates the deviation from the Tm
d
value of the corresponding full-match triplex. Values in parentheses indicate the Tm at pH 7.0. Each Tm value is the average of three separate
experiments.
temperature range compared to that of the triplex formed with an
unmodified parental chimera (GK-300). The Tm values of the
triplex estimated from the first derivative of the UV-melting
profiles are listed in Table 1.
alkyl linker as well as to create novel functional chimera TFOs
utilizing the current modification methodology is now underway in
our laboratory and will be reported elsewhere.
The work was partially supported by a Grant-in-Aid for
Scientific Research from the Ministry of Education, Culture,
Sports, Science, and Technology, Japan. K. S. also thanks the
As shown in Table 1, the modified TFO effectively stabilizes the
triplex since the Tm values of the triplexes consisting of the modified
TFO and full-matched target (ODN-2 : ODN-3) are considerably
higher (> 10 uC) than that of the triplex containing unmodified GK-
300. The triplex-stabilizing effect is, however, dependent on the
length of the alkyl linker connecting the anthraquinone moiety and
the propionic acid moiety in the incorporated multi-conjugate. Thus,
the TFOs bearing a longer alkyl linker (GK-354A and -354B)
exhibited higher Tm values compared to the TFOs bearing a shorter
alkyl linker (GK-358A and -358B). It is noteworthy that the Tm
values of GK-354A and -354B are higher than the physiological
temperature even at pH 7.0.
Ministry for supporting A. Z. A. as
a scholar of the
Monbukagakusho Scholarship for Foreign Graduate Students.
Notes and references
31P NMR (CDCl3, d); 148.6 ppm (singlet) (n = 2); 148.4 ppm (singlet)
{
(n = 6).
§ The average molecular weights of the modified chimera TFOs calculated
from the multiply charged ion peaks were as follows: GK-300, 7824.91
(calcd 7825.25); GK-354A, 7868.55; GK-354B, 7868.32 (calcd 7867.46) and
GK-358A, 7813.19; GK-358B, 7813.29 (calcd 7811.40).
Interestingly, the data in Table 1 also revealed that the difference
in the Tm values between the two diastereomers was quite small in
all cases (, 1 uC). This small difference in the Tm values between
the diastereomers indicates that the environment around the linker
portion of the modified TFOs is very flexible. It also suggests that
it is highly versatile for the modification at the linker portion of the
chimera TFO with this type of non-nucleosidic unit to functiona-
lize it without considering the stereochemistry in the modification.
As a result, both GK-354A and -354B stabilized an alternate-
stranded triplex at more than 15 uC with even a single modification
of the parental chimera TFO.
1 (a) C. He´le`ne, Anti-Cancer Drug Des., 1991, 6, 569; (b) I. Luyten and
P. Herdewijn, Eur. J. Med. Chem., 1998, 33, 515; (c) D. Praseuth,
A. L. Guieysse and C. He´le`ne, Biochim. Biophys. Acta, 1999, 1489, 181;
(d) D. M. Growers and K. R. Fox, Nucleic Acids Res., 1999, 27, 1569;
(e) M. Grigoriev, D. Praseuth, A. L. Guieysse, P. Robin, N. T. Thuong,
C. He´le`ne and A. Harel-Bellan, Proc. Natl. Acad. Sci. U. S. A., 1993, 90,
3501.
2 D. A. Horne and P. B. Dervan, J. Am. Chem. Soc., 1990, 112, 2435.
3 (a) L. C. Grifiin and P. B. Dervan, Science, 1989, 245, 967; (b)
J. A. Fossella, Y.-J. Kim, H. Shih, E. G. Richard and J. R. Fresco,
Nucleic Acids Res., 1993, 21, 4511; (c) J.-L. Mergny, J.-S. Sun,
M. Rouge´e, T. Garestier, F. Barcelo, J. Chomilier and C. He´le`ne,
Biochemistry, 1991, 30, 9791; (d) A. Ono, C.-N. Chen and L.-S. Kan,
Biochemistry, 1991, 30, 9914; (e) G. C. Best and P. B. Dervan, J. Am.
Chem. Soc., 1995, 117, 1187; (f) W. A. Greenberg and P. B. Dervan,
J. Am. Chem. Soc., 1995, 117, 5016; (g) D. M. Gowers and K. R. Fox,
Nucleic Acids Res., 1997, 25, 3787; (h) Y. Ueno, M. Mikawa, S. Hoshika
and A. Matsuda, Bioconjugate Chem., 2001, 12, 635; (i) S. Hishika,
Y. Ueno and A. Matsuda, Bioconjugate Chem., 2003, 14, 607; (j)
S. Hoshika, Y. Ueno, H. Kamiya and A. Matsuda, Bioorg. Med. Chem.
Lett., 2004, 14, 3333.
In addition to these results, the Tm values of the triplexes
consisting of the modified TFO and the mismatched target
(ODN-4 : ODN-5) are significantly reduced in Table 1. The
magnitude of the Tm decrement is nearly the same as that of the
parental unmodified GK-300 at both pH 6.5 and 7.0. The results
clearly indicate that the modified TFOs retain almost the same
degree of sequence discrimination ability as the unmodified TFO
and the introduction of the intercalator–polyamine conjugate to
the chimeric TFO would not abolish the sequence discrimination
ability despite its apparent enhancement effect on the thermal
stability with a full-matched target.
4 K. Shinozuka, N. Matsumoto, H. Suzuki, T. Moriguchi and H. Sawai,
Chem. Commun., 2002, 2712.
5 T. Miyashita, N. Matsumoto, T. Moriguchi and K. Shinozuka,
Tetrahedron Lett., 2003, 44, 7399.
6 A. T. M. Z. Azam, M. Hasegawa, T. Moriguchi and K. Shinozuka,
Bioorg. Med. Chem. Lett., 2004, 14, 5747.
7 (a) H. Asanuma, T. Ito and M. Komiyama, Tetrahedron Lett., 1998, 39,
9015; (b) X. Liang, H. Asanuma and M. Komiyama, Tetrahedron Lett.,
2001, 42, 6723.
8 N. D. Shinha, J. Biernat and H. Koster, Tetrahedron Lett., 1983, 24, 5843.
9 Y. Sato, G. Tateno, K. Seio and M. Sekine, Tetrahedron Lett., 2002, 43,
3251.
In conclusion, we were able to successfully synthesize a multi-
conjugate of 2,2-bis(hydroxymethyl)propionic acid–anthraqui-
none–polyamine and incorporate it into the a–b chimera TFO.
The modification was quite effective at enhancing the triplex-
forming ability of the TFO without abolishing sequence specificity.
The enhancement effect depends on the length of the alkyl linker,
however, it was almost independent of the stereochemistry of the
incorporated multi-conjugate. Work to optimize the length of the
10 W. L. Sung, Chem. Commun., 1981, 1089.
11 G. R. Gough, M. J. Brunden and P. T. Gilham, Tetrahedron Lett., 1981,
22, 4177.
12 See the Electronic Supplementary Information (ESI).
This journal is ß The Royal Society of Chemistry 2006
Chem. Commun., 2006, 335–337 | 337