The spermine-bisaryl conjugate as a potent inducer of B- to Z-DNA transition

Article abstract of DOI:10.1002/chem.201000947

DNA containing alternating purine and pyrimidine repeats has the potential to adopt the Z-DNA structure, one of the well-studied structures besides A- and B-DNA. Despite a number of molecular models that have been proposed to explain the mechanism for B→Z

Full text of DOI:10.1002/chem.201000947

FULL PAPER
DOI: 10.1002/chem.201000947
The Spermine–Bisaryl Conjugate as a Potent Inducer of
B- to Z-DNA Transition
AHCTUNGTRENNUNG
Issei Doi, Genichiro Tsuji, Kyoko Kawakami, Osamu Nakagawa, Yosuke Taniguchi, and
Shigeki Sasaki*[a]
Abstract: DNA containing alternating
purine and pyrimidine repeats has the
potential to adopt the Z-DNA struc-
ture, one of the well-studied structures
besides A- and B-DNA. Despite a
number of molecular models that have
been proposed to explain the mecha-
nism for B!Z transition, there is con-
tinued discussion on the mechanism
and physiological role of this transition.
In this study, we have found that the
bis(2-naphthyl)-maleimide–spermine
conjugate (3c) exhibits a remarkable
ability to cause the B!Z transition of
thalpically and entropically favorable.
The ligand might effect the dehydra-
tion of B-DNA, which leads to the B!
dACTHNUTRGEN(NUG CGCGCG)2 at low salt concentra-
tions. Using isothermal titration calo-
rimetry (ITC) we show that the B!Z
transition induced by 3c is both en-
Z transition. Interestingly, an inter-
mediate CD between the B and Z
forms was observed in the pH-depen-
dent transition in the presence of the
ligand. The unique structure and char-
acteristics of the ligand designed in this
investigation will be useful for the
study of Z-DNA.
Keywords: B–Z transition · circular
dichroism
titration calorimetry
plasmon resonance
·
DNA
·
isothermal
· surface
Introduction
ACTHNGUTERNNUG
presence of transition metal complexes[7] ([Co(NH3)6]3+), or
small molecules, such as polycations (spermine), chiral bind-
ing molecules, etc.[8,9] The B!Z transition is cooperative,
and the length of the polymer is important for the process,
DNA containing alternating purine and pyrimidine repeats
has the potential to adopt the Z structure, one of the well-
studied structures besides A- and B-DNA. Z-DNA is a left-
handed helix in which the purines are in the syn conforma-
tion and the pyrimidines in the anti conformation.[1] It has
been shown that Z-DNA exists in vivo under physiological
conditions as a transient structure occasionally induced by a
biological process, such as transcription, the methylation of
cytosine, and the level of DNA supercoiling.[2,3] Several pro-
teins have been identified that bind to Z-DNA.[2] Recently,
the B!Z transition has been utilized as a key event in
nanodevices that are responsive to an external signal.[4–6]
The B!Z transition is observed under a variety of condi-
tions including high salt concentrations (4m NaCl), in the
for instance, the B form of poly[dACHTUNTRGENN(UG G-C)·dACHTUNGTERN(NUGN C-G)] is more
easily transformed into the Z form with increasing polymer
length.[10,11] Despite a number of molecular models that
have been proposed for the mechanism and the physiologi-
cal role of the B!Z transition, there is continued discus-
sion. We now report that spermine-conjugated bisaryl li-
gands (Scheme 1) effectively induce the B!Z transition of
dACTHNUTRGNEN(UG CGCGCG)2 at low salt concentrations.
[a] I. Doi, G. Tsuji, K. Kawakami, Dr. O. Nakagawa, Dr. Y. Taniguchi,
Prof. S. Sasaki
Graduate School of Pharmaceutical Sciences, Kyushu University
3-1-1 Maidashi, Higashi-ku, Fukuoka, 812-8582 (Japan)
Fax : (+81)92-642-6615
Scheme 1. Structural characteristics of the topology of: A) Z-DNA, and
B) a representative ligand (3c) synthesized in this study. Because of the
syn conformation of guanosine, the minor groove of Z-DNA is narrower
and the phosphate anions are closer than those of B-DNA.
E-mail: sasaki@phar.kyushu-u.ac.jp
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201000947.
Chem. Eur. J. 2010, 16, 11993 – 11999
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
11993

Results and Discussion
In the unique topology of Z-DNA, the guanine bases in the
syn conformation base pair with the cytosine bases in the
anti conformation, and the minor groove is narrower and
deeper than that of B-DNA.[1] It has been reported that in a
low temperature crystal structure of the spermine com-
plexed with Z-DNA duplex, the spermine in the minor
groove decreases the cross-groove electrostatic repulsions
between the phosphate anions within the Z-DNA.[12] The
crystal structure has been reported to bind two molecules of
tetra-amine, NH2ACHTUNGTRENNUNG((CH2)3NH)3H, in the minor groove of Z-
d
ACHTUNGTRENNUNG
During the initial stage of this study, we hypothesized that
spermine conjugated with aromatic groups would produce
electrostatic interactions between the cations of spermine
and phosphate anions within the minor groove of Z-DNA.
Based on the crystal structure in which the spermine is lo-
cated in the minor groove of Z-DNA, it was also expected
that the aromatic groups would provide additional stacking
interactions. Because the space between the G–C base pair
of Z-DNA is narrower than that of B-DNA, the bisaryl
structure, as shown in Scheme 1B, was postulated to be a
better candidate for a good fit for such a space in the Z-
DNA.
Scheme 2. Synthesis of bisaryl ligands and the structure of the reference
compound 5.
companied by an increase in intensity of the negative band
at about 296 nm and the positive band at about 270 nm, and
clearly indicates B!Z transition. The B!Z transition
reached equilibrium immediately after the addition of the
bisaryl ligand. The concentrations of ligand needed to
induce a 50% B!Z transition (EC50) are summarized in
Table 1. As clearly shown by the EC50 values, the aromatic
group is a key determinant for the B!Z transition ability.
The structure of the designed compounds and their syn-
theses are summarized in
Scheme 2. The dibromomalei-
mide derivative (1) was coupled
to the aromatic compounds,
and the resulting bisaryl deriva-
tives (2) were transformed into
the corresponding anhydrides.
Aminolysis of the anhydride
with the amine was followed by
acid-catalyzed imide closure to
produce the bisaryl-maleini-
mide-amine derivatives (3 and
4). They were used as the HCl
or TFA salts. The spermine–an-
thracene conjugate (5) has been
reported to bind poly[dACTHUNTGRENNG(U G-C)·d-
ACHTUNGTRENNUNG(C-G)] by anthracene intercala-
tion,[14] and was used as a refer-
ence compound.
The B!Z transition was ana-
lyzed by CD spectrum changes
by using dACHTUNGTRENNUNG(CGCGCG)2 as the
B-DNA substrate. Examples
with 3a, 3c and 3d are shown
in Figure 1A, B and C, respec-
tively. Addition of the bisaryl
ligand decreased the intensity
of the positive (at about
278 nm) and negative bands (at
about 253 nm) that are charac-
Figure 1. CD change in B-DNA after the addition of the bisaryl ligands. A) B!Z transition induced by 3a,
B) B!Z transition induced by 3c, C) B!Z transition induced by 3d, D) pH-dependent B!Z transition. A),
B), C) The CD spectrum was measured by using B-DNA dACHTNURGTNEUNG(CGCGCG)2 (20 mm) in Na cacodylate buffer
(5 mm) containing NaCl (100 mm) at pH 7.0. Ligand concentrations: A) 0 to 650 mm, B) 0 to 60 mm, C) 0 to
409 mm, D) a mixture containing 20 mm d(CGCGCG)2 and 60 mm 3c in Na cacodylate (5 mm) and NaCl
AHCTUNGTRENNUNG
teristic of B-DNA; this was ac- (100 mm) was used. The pH was raised from 7 to 9.4 by the addition of NaOH (0.1m).
11994
www.chemeurj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 11993 – 11999

Spermine–Bisaryl Induces B- to Z-DNA Transition
FULL PAPER
Table 1. Properties of the bisaryl ligands 3 and 4.
To determine the role of the spermine-conjugated bisaryl
ligand in the B!Z transition, the binding properties of the
ligand were investigated. Because 8-methyl-2’-deoxyguano-
sine (m8G) is known to stabilize the Z form when it is incor-
[c]
Ligand
EC50 [mm][a] DTm [oC] Z-DNA[b]
Ka [106 mꢀ1
]
Z-DNA B-DNA
3a
3b
3c
3d
3e
4a
250
>600
18
88
29
370
7.0
3.8
0.28
–
1.1
–
porated into a CG repeat sequence, dACTHNUTRGNEUNG
(CGCm8GCG)2 was
18
6.0
3.2
0.04
0.7
0.26
used as the standard Z-DNA.[17] The effect of the ligand on
the thermal stability of the Z-DNA was estimated by meas-
uring the UV melting temperature (Tm); the increase in the
Tm values are summarized in Table 1. The ligand 3c showed
a high stabilizing effect with Z-DNA. However, because
there is not a certain relationship between the ED50 and
DTm values, the increase in the thermal stability of the Z-
DNA in the presence of ligand is not thought to be the
cause of the B!Z transition.
[e]
[e]
[e]
–
–
–
–
–
–
–
–
–
–
4b
4c
spermine
5
>600
>600
1.3
0.5
5.1
–
no[d]
0.12
0.11
0.01[f]
0.08[g]
0.24
0.052
0.012[f]
0.014[g]
no[d]
(ꢀ)-daunorubicin
–
–
–
–
helicene (P)-A
[a] The EC50 values for the B!Z transition were obtained from the CD
experiments. [b] Melting temperatures were measured by using d-
Next, we investigated the binding properties of the ligand
with SPR measurements. The 5’-biotin-conjugated
d(TTTTCGCGCG-TTTT-CGCGCG) sequence was immo-
bilized on a sensor chip for measurement of ligand bind-
ing.[18] This sequence formed the B-type duplex with a loop
structure, but was not transformed to the Z form after addi-
tion of 3c under low salt conditions (Figure S9C in the Sup-
ACHTUNGTRENNUNG
(CGCm8GCG)2 (5 mm) and ligand (20 mm) in Na cacodylate buffer
(5 mm) containing NaCl (100 mm) at pH 7.0. [c] The binding constants
were obtained by SPR measurements by using sensor chips immobilized
with 5’-biotin-conjugated d(TTTTCGCGCG-TTTT-CGCGCG) as the B-
DNA
and
5’-biotin-conjugated
d(TTTTCGCm8GCG-TTTT-
CGCm8GCG) as the Z-DNA. [d] No CD change was observed at 600 mm.
[e] Measurements were not done because of the overlapping UV absorp-
tion of 3e. [f] Data taken from ref. [8]. [g] Data taken from ref. [9].
porting
Information).
However,
5’-biotin-conjugated
d(TTTTCGCm8GCG-TTTT-CGCm8GCG) formed the Z-
DNA under the same conditions (Figure S9D in the Sup-
porting Information). Accordingly, the binding properties
for the B- and Z-DNA were independently measured by
SPR with these duplexes. Examples of the SPR results with
3c and B- and Z-DNA are shown in Figure 2A and B, re-
spectively. The SPR data (Figure S10 in the Supporting In-
The effect for the B!Z transition is in the order 3c>3e>
3d>3a>3b among the spermine conjugates. The anthra-
cene–spermine conjugate (5), a potent B-DNA binder,[14]
did not induce B!Z transition (Figure S9A and B in the
Supporting Information). The bis(2-naphthyl) derivative
(3c) was found to have the highest efficacy among the li-
gands tested in this study.
The amino group also has a significant effect on the effi-
ciency of B!Z transition, and spermine produced the best
effect compared to spermidine (4a), tris(2-aminoethyl)-
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
DNA at the low salt concentrations used in this study (Fig-
ure S6 in the Supporting Information). The bisaryl–spermine
ligands are unique examples of small molecules demonstrat-
ing a high efficiency in inducing B!Z transition. The Z
form returned to the B form when the pH was raised from
7.0 to 9.4 (Figure 1D), indicating the importance of the elec-
trostatic binding between the ammonium cations and the
phosphate anions for the B!Z transition. Interestingly, Fig-
ure 1D shows two isodichroic points at around 283 and
267 nm, and the existence of an intermediate CD is suggest-
ed in addition to the B and Z forms. The intermediate CD
resembles that of the B form. The Z form changed to the in-
termediate form when the pH was raised from 7.0 to about
8.1, and the intermediate form returned to the B form at
higher pH. From the pKa values of the carbamate conjugate
of spermine (10.1, 8.6, 7.3),[15] the bisaryl ligand 3c with the
decreased positive charge should bind the intermediate com-
plex in the B!Z transition. This intermediate CD is similar
to that observed during the slow B!Z transition of poly[d-
Figure 2. SPR sensorgrams obtained by using an ODN with a loop struc-
ture in the B and Z forms. A) Transition to B-DNA, B) transition to Z-
DNA. The biotin-conjugated ODN shown in the graph was immobilized
on an avidin-coated sensor chip and 3c (0.5–1.6 mm) was analyzed for
5 min.
N
ACHTUNGTRENNUNG(C-G)] induced by the cationic comb-type copoly-
Chem. Eur. J. 2010, 16, 11993 – 11999
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
11995

S. Sasaki et al.
Table 2. Thermodynamic parameters obtained by the titration of 3c
formation) were analyzed by using the standard method in-
stalled on the instrument to give the equilibrium association
constants (Ka), which are summarized in Table 1. Among
the tested ligands, only 3c showed a higher binding affinity
for the Z-DNA than B-DNA. In this case, the bisaryl-malei-
mide part increased the binding affinity to Z-DNA by 20–30
times compared to spermine; this suggests an important role
for the geometry of the ligand in DNA binding. Although
3a and 3d induced B!Z transition with moderate potency,
their binding affinities to Z-DNA were relatively low when
compared to B-DNA. Accordingly, the binding affinity of
the ligand to Z-DNA could not be correlated with the po-
tency for the B!Z transition.
against B- and Z-DNA.[a]
Inducer
Ka
[106mꢀ1
H
A
G
A
ꢀTS
[kJmolꢀ1
]
N[d]
G
]
[kJmolꢀ1
]
[kJmolꢀ1
]
U
B–3c<10 mm
B–3c>10 mm
Z–3c
1.23
0.62
3.08
0.0345
0.0002
3.34
ꢀ77.23
ꢀ76.05
ꢀ72.23
ꢀ25.72
ꢀ11.72
ꢀ18.00
ꢀ80.57
ꢀ54.28
ꢀ39.58
ꢀ27.57
ꢀ15.03
ꢀ20.93
1.07
3.5
4.06
6.58
5.3
ꢀ21.77
ꢀ32.56
1.85
2.93
3.01
spermine[b]
Na+[c]
[Co
(NH3)6]3+[c] 0.1
5.1
[a] Thermodynamic parameters were obtained with the curve-fitting
method by using standard procedures, two independent models for titra-
tion with the B-DNA and one independent model for titration with the
Z-DNA. [b] The titration of spermine was done against d(GC)20 and did
not induce B!Z transition.[19] [c] The titration was done against (dm5C-
dC)4 and induced B!Z transition; the parameters are reported per base-
pair.[20] [d] The binding stoichiometry is defined as molar ratio [ligand]/
[duplex].
To obtain further insights into the origin of the potency of
3c to induce B!Z transition, thermodynamic parameters
were obtained by ITC. Ligand was added into a solution
containing d
in the Z form. Because 3c induces the B!Z transition of d-
(CGCGCG)2 under the titration conditions, the parameters
(CGCm8GCG)2
ACHTUNGTRENNUNG(CGCGCG)2 in the B form or dHCATUNGTRENNUGN
G
likely that the initial binding occurs with the spermine part
of 3c. However, the second step involving the binding and
B!Z transition was shown to be both enthalpically as well
as entropically favorable. This is a characteristic feature of
B!Z transition induced by 3c compared to the well-known
obtained by using the B form include those for the B!Z
transition. Figure 3 shows the heats obtained from the titra-
tion of 3c into B- or Z-DNA. The isotherm shown at the
bottom of Figure 3A suggests that the initial step with less
than 10 mm is different to when higher concentrations of
ligand are used, as highlighted by the dotted square. Such a
change in the isotherm was confirmed by a titration with the
ligand at low concentrations (Figure 3A, inset). The thermo-
dynamic parameters obtained for 3c are shown in Table 2
together with the reported ones. The thermodynamic param-
eters obtained with 3c and B-DNA clearly indicate that the
initial binding step is entropy driven, and are similar to
those obtained with spermine in that the binding of sper-
B!Z transition induced by [Co
(NH3)6]3+ or Na+, which are
ACHTUNGTRENNUNG
entropy-driven processes[20] (Table 2). The binding stoichi-
ometry defined as the molar ratio was 1.07 below 10 mm and
changed to 3.5 at higher concentrations. The isotherm ob-
tained with Z-DNA indicated one-phase binding and is fa-
vorable both entropically and enthalpically. The stoichiome-
try was shown to be Z-DNA/ligand 3c=1:4.0.
The complex structure formed with 3a and d-
1
N
mine to d
(C-G)20, which is not B!Z inducible, occurs with
spectroscopy. The ligand 3a was used because of its high sol-
ubility in the buffer for the 1H NMR measurements. The
NMR spectra of free 3a, free
Z-DNA, and the complex are
compared in Figure 4. All pro-
tons of the indolyl groups were
shifted upfield (Figure 4, A vs.
B). Some signals of the Z-DNA
(8-H of G6, and 6-H of C)[15]
also displayed upfield shifts by
the formed complex (Figure 4,
B vs. C), suggesting that the in-
dolyl groups were stacked with
the G–C base pairs at the ter-
minal position. Thus, the com-
plex can be postulated to have
this structure as the bisaryl
parts come close to the G–C
pairs at the terminal position
a small positive DH and a larger ꢀTDS (Table 2).[19] It is
and the cationic spermine inter-
acts with the phosphate anions
in the minor groove and/or at
the exterior of the duplex.
Figure 3. Unsubtracted calorimetric binding heats obtained from titration of ligand 3c against B-DNA d-
(CGCmGCG)2 (B). The positive (endothermic) heats show the ligand–DNA
ACHTUNGTRENNUNG(CGCGCG)2 (A) and Z-DNA dAHCTUNGTRENNGUN
binding, and the negative (exothermic) heats represent the ligand heat of dilution. Subtraction of the heats of
dilution obtained from the separate titrations from the complete titration curve yielded the binding isotherm
reported at the bottom. The ligand solution was added to the solution containing 10 mm B-DNA or Z-DNA in
Na cacodylate buffer (5 mm) containing NaCl (100 mm) at pH 7.0. The ligand concentration ranged from 0 to
148 mm. The inset shows the isotherm obtained when the ligand concentration ranged from 0–15.3 mm.
11996
www.chemeurj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 11993 – 11999

Spermine–Bisaryl Induces B- to Z-DNA Transition
FULL PAPER
Such a stable intermediate form might allow structure deter-
mination, which will be useful to gain insight into the molec-
ular mechanism underlying the B!Z transition.
Conclusion
In conclusion, we have found that the bis(2-naphthyl)malei-
mide–spermine conjugate (3c) exhibits a remarkable ability
for the induction of the B!Z transition of dACHTUGNRTEN(NUNG CGCGCG)2 at
low salt concentrations. The ITC study has revealed that the
B!Z transition induced by 3c is both enthalpically and en-
tropically favorable. The ligand may effect the dehydration
of B-DNA that leads to the B!Z transition. Interestingly,
an intermediate CD between the B and Z forms was ob-
served in the pH-dependent transition in the presence of the
ligand. The unique structure and characteristics of the
ligand designed here will be useful for the study of Z-DNA.
Experimental Section
DNA oligomers: DNA oligomers were synthesized with an automated
DNA synthesizer by using conventional amidite chemistry. The DMTr-
protected DNA was cleaved from the resin with NH4OH solution and pu-
rified by HPLC equipped with a ODS column by using TEAA·CH3CN
(0.1m) linear gradient system. The DMTr protecting group of the purified
oligonucleotide (ODN) was cleaved in aqueous AcOH (10%) and the
mixture was washed with ether. The purity of the ODN was confirmed
by MADI-TOF MS and UV measurements. The concentration of the
stock solution was estimated by UV absorbance at 260 nm, which was fur-
ther confirmed by HPLC analysis of the component nucleosides after en-
zymatic hydrolysis with bacterial alkaline phosphatase (BAP) and snake
venom phosphodiesterase (VPDE), with less than 5% error.
Figure 4. 1H NMR spectra of the complex of 3a with Z-DNA. A) Ligand
3a only, B) the complex of 3a and Z-DNA, C) Z-DNA only. The marked
peaks correspond to the coexisting B-DNA.[17] The 1H NMR spectra was
obtained by using 120 mm Z-DNA and 100 mm 3a in D2O containing Na
cacodylate (5 mm), NaCl (100 mm) at pH 7 and 108C.
Discussion
CD measurements: CD spectra were recorded by using a Jasco-J720
spectrometer with a 2 mm quartz glass cell (200 mL) at the constant tem-
The B!Z transition is associated with nucleotide dehydra-
tion.[21–23] It has been also reported that neutral solutes facili-
tate the B!Z transition through osmotic stress.[24,25] The
thermodynamic parameters of 3c obtained by ITC have sug-
gested that an entropy-driven initial binding of the spermine
part is followed by an entropically and enthalpically favora-
ble process. The NMR spectroscopy study has shown that
the aromatic groups assemble around the terminal G–C
pairs. Thus, in the second step, which is an enthalpically and
entropically favorable process, association of the hydropho-
bic bisaryl groups might cause dehydration of the B-DNA,
and enhance electrostatic binding of the spermine part to
the phosphate backbone of the DNA, eventually leading to
the B!Z transition. The geometry of the bisaryl group is
also a significant determinant for the potency in the B!Z
transition. Taking into account that some intercalators inhib-
it B!Z transition,[26–29] it should be noted that the 2-naph-
thyl groups of 3c are twisted towards each other and are not
likely to intercalate between the base pairs. The role of the
bisaryl geometry for the B!Z transition will be further
studied in detail. It is also notable that the intermediate CD
induced by 3c in the pH-dependent B!Z transition (Fig-
ure 1D) was stable under the conditions used in this study.
perature of 208C. A solution of d
AHCTUNGTRENNUNG
ACHTUNGTERN(NUNG CGCGCG)2 B-DNA (20 mm) or d-
ing 100 mm NaCl at pH 7) was used for the CD measurements. The ODN
solution was titrated with the ligand stock solution (1 mL each). All CD
data were blank corrected, and are reported as the average of quadrupli-
cate scans.
Measurement of UV melting temperature: In the UV melting experi-
ments, a solution of Z-DNA (5 mm in Na cacodylate buffer 5 mm contain-
ing 100 mm NaCl, pH 7) was used. The temperature was raised or low-
ered at the rate of 18Cminꢀ1 between 10 and 608C.
Surface plasmon resonance (SPR) measurements: 5’-Biotin-conjugated
d(TTTTCGCGCG-TTTT-CGCGCG) or d(TTTTCGC-m8GCG-TTTT-
CGC-m8GCG) (10–500 nm) in HEPES (10 mm) and NaCl (0.15m) at
pH 7.4 was immobilized on a sensor chip SA at 208C. One flow channel
was always left as a blank for reference. The SPR measurements were
performed at 258C by using Na cacodylate buffer (5 mm) containing
NaCl (100 mm) at pH 7.0 as the running buffer. Binding was measured at
20 mLminꢀ1 for 120 s and dissociation for 150 s. A Biacore 3000 (GE
Healthcare) was used for the SPR measurements.
Isothermal titration calorimetry (ITC): The ligand solution (3c; 1 mm)
was added to the final dialysate obtained from ODN dialysis, and was ti-
(CGCm8GCG)2 in Na cacodylate
trated into 10.0 mm dACTHNUTRGENN(GU CGCGCG)2 or dACHTUGNTRENNUGN
buffer (5 mm) containing NaCl (100 mm) at pH 7.0 and 25.08C. The first
injection was 2 mL, and then 10 mL of each ligand stock was added to the
ODN solution. The data obtained with the first injection were omitted
from the analysis. The ODN concentrations are reported as molar con-
centrations of double strands. All samples were degassed prior to use by
Chem. Eur. J. 2010, 16, 11993 – 11999
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
11997

S. Sasaki et al.
application of vacuum and simultaneously stirred on a magnetic plate. A
NanoITC SV (TA Instruments) was used for the calorimetric measure-
ments.
1H NMR measurements: 1H NMR spectra were recorded by using a
3286, 3057, 2931, 2861, 1766, 1698, 1443, 1405 cmꢀ1; HRMS (ESI): m/z:
calcd for C26H35N4O2 [M+H]+: 435.2755; found 435.2749.
Compound 3d: 1H NMR (400 MHz, CD3OD): d=1.73 (4H, brs), 1.96–
2.04 (2H, m), 2.11–2.16 (2H, m), 2.99–3.08 (8H, m), 3.18 (2H, t, J=
7.6 Hz), 3.84 (2H, t, J=6.5 Hz), 7.26 (2H, t, J=7.8 Hz), 7.37 (5H, brs),
7.61 (1H, br), 7.59 (2H, d, J=7.6 Hz), 7.77–7.84 ppm (4H, brs);
13C NMR (125 MHz, CD3OD): d=172.3, 141.7, 134.9, 132.0, 131.2, 129.6,
129.4, 128.6, 127.2, 127.1, 126.5, 126.0, 48.2, 46.8, 45.8, 37.8, 36.5, 26.9,
25.3, 24.2, 24.2 ppm; IR: n˜ =3045, 2838, 1768, 1702, 1675, 1200,
1130 cmꢀ1; HRMS (ESI): m/z: calcd for C34H39N4O2 [M+H]+: 535.3068;
found 535.3062.
Varian UNITY INOVA 600 MHz spectrometer with 3a (100 mm) and d-
ACHTUNGTRENNUNG
(CGCm8GCG)2 (120 mm) in D2O containing Na cacodylate (5 mm) and
1
NaCl (100 mm) at 108C. Each H NMR spectra of 3a or dACTHNUTRGNEUNG
(CGCm8GCG)2
was recorded under the same conditions.
The synthesis of 3c as a general procedure: A solution of CsF (0.53 g,
3.48 mmol, in 1.5 mL H2O) was added to a mixture of 1-benzyl-3,4-dibro-
mopyrole-2,5-dione (0.30 g, 0.87 mmol), 2-naphthalene boronic acid
Compound 3e: 1H NMR (400 MHz, CD3OD): d=1.82–1.84 (4H, brs),
2.00–2.07 (2H, m), 2.26–2.30 (2H, m), 3.00 (2H, t, J=7.8 Hz), 3.05–3.08
(4H, m), 4.03 (2H, t, J=6.4 Hz), 7.73 (2H, d, J=9.2 Hz), 7.89–8.11 ppm
(16H, m); 13C NMR (125 MHz, CD3OD): d=172.6, 141.9, 133.6, 132.1,
131.6, 130.3, 129.6, 129.2, 128.7, 128.5, 128.3, 127.9, 127.1, 126.8, 126.6,
125.6, 125.4, 125.1, 48.9, 48.2, 46.9, 45.8, 37.8, 36.7, 27.1, 25.3, 24.3,
24.2 ppm; IR: n˜ =3033, 2840, 1764, 1704, 1679 cmꢀ1; HRMS (ESI ): m/z:
calcd for C46H42N4O2 [M+H]+, 683.3381; found 683.3388.
(0.45 g, 2.61 mmol), PdACHTUNGTRNEUGN(PPh3)2Cl2 (30.50 mg, 0.0435 mmol) and
PhCH2NEt3Cl (10 mg, 0.0435 mmol) in degassed toluene–H2O (3.6 mL/
1.8 mL), and was heated at 608C for 30 h. The reaction mixture was dilut-
ed with AcOEt (15 mL), and washed successively with HCl (1m; 15 mL)
and brine. The organic layers were dried over Na2SO4, and evaporated to
give a crude residue, which was purified by column chromatography
(silica gel, hexane/AcOEt, 7:1) to give 1-benzyl-3,4-bis(2-naphthyl)-pyr-
role-2,5-dione (0.38 g, 99%) as
a
yellow solid. 1H NMR (400 MHz,
Compound 4a: 1H NMR (400 MHz, CD3OD): d=1.50–1.59 (4H, m),
1.90 (2H, q, J=6.9 Hz), 2.59 (2H, t, J=7.0 Hz), 2.65 (4H, dt, J=3.4,
6.7 Hz), 3.72 (2H, d, J=6.7 Hz), 6.59 (2H, dt, J=0.9, 8.2 Hz), 6.83 (2H,
d, J=8.2 Hz), 6.97 (2H, dt, J=0.9, 8.2 Hz), 7.32 (2H, d, J=8.2 Hz),
7.73 ppm (2H, s); 13C NMR (125 MHz, CD3OD): d=174.1, 137.8, 130.1,
128.6, 127.0, 123.1, 122.5, 120.7, 112.5, 107.5, 48.2, 46.8, 40.0, 35.8, 25.5,
24.2, 19.3 ppm; IR: n˜ =3373, 2928, 2857, 2345, 1753, 1688, 1614, 1529,
1436 cmꢀ1; MS (ESI): m/z: calcd for C27H29N5O2 [M+H]+: 456.2394;
found 456.2390.
Compound 4b: 1H NMR (400 MHz, CD3OD): d=2.96 (6H, t, J=
5.2 Hz), 3.11 (4H, t, J=6.7 Hz), 3.81 (2H, t, J=5.8 Hz), 6.61 (2H, dd, J=
7.3, 7.7 Hz), 6.85 (2H, d, J=7.7 Hz), 6.97 (2H, dd, J=7.3, 7.9 Hz), 7.34
(2H, d, J=7.9 Hz), 7.77 ppm (2H, s); 13C NMR (125 MHz, CD3OD): d=
173.9, 137.8, 130.1, 128.9, 127.0, 123.1, 122.5, 120.7, 112.4, 107.6, 49.1,
40.3, 36.7 ppm; IR: n˜ =3385, 2924, 2356, 1755, 1693, 1614, 1530,
1417 cmꢀ1; HRMS (ESI): m/z: calcd for C26H29N6O2 [M+H]+: 457.2347;
found 457.2339.
CDCl3): d=4.86 (2H, s), 7.27–7.37 (5H, m), 7.46–7.53 (6H, m), 7.67 (2H,
d, J=8.6 Hz), 7.77 (2H, d, J=7.9 Hz), 7.81 (2H, d, J=7.9 Hz), 8.21 ppm
(2H, s); IR: n˜ =3675, 2973, 1766,1702, 1401 cmꢀ1
[M+H]+: 440.10.
; MS (ESI): m/z:
The above compound (0.15 g, 0.35 mmol) was dissolved in 5m NaOH
(2.2 mL)–ethanol (9.1 mL), and the solution was stirred for 21 h at room
temperature. The reaction mixture was acidified with HCl (1m; 12 mL)
and extracted with CHCl3. The organic layers were dried over Na2SO4,
and evaporated to give a crude residue, which was purified by column
chromatography (silica gel, toluene) to give 3,4-bis(2-naphthyl)furan-2,5-
dione as a yellow solid (0.113 g, 93%). 1H NMR (400 MHz, CDCl3): d=
7.42 (2H, d, J=8.6), 7.51–7.59 (4H, m), 7.57 (2H, dd, J=7.3 Hz,), 7.74
(2H, d, J=8.6 Hz), 7.82 (2H, d, J=7.9 Hz), 7.86 (2H, d, J=8.2 Hz),
8.29 ppm (2H, s); IR: n˜ =1839, 1762, 865, 822, 738 cmꢀ1; MS (ESI): m/z:
[M+H]+: 351.21.
A solution of the above compound (16.6 mg, 0.051 mmol, in 0.5 mL
ether) was added slowly into a solution of spermine (22 mg, 0.152 mmol,
in 3 mL methanol), and the mixture was stirred for 1 h at room tempera-
ture. The precipitates were collected and dissolved in toluene (0.75 mL).
The solution was heated under reflux for 19 h, and the solvents were re-
moved to give a residue, which was purified by column chromatography
(amino-silica gel, CHCl3/MeOH, 100:0 to 5:1) to give 3c as a red solid.
This material was transformed to the corresponding hydrochloride with
HCl·CH3OH (36 mg, 15%).
Compound 4c: 1H NMR (400 MHz, CD3OD): d=3.21 (2H, brs), 3.99
(2H, t, J=5.6 Hz), 6.60 (2H, dd, J=7.3, 7.9 Hz), 6.60 (2H, d, J=7.9 Hz),
6.97 (2H, dd, J=7.3, 7.9 Hz), 7.33 (2H, d, J=7.9 Hz), 7.78 ppm (2H, s);
13C NMR (125 MHz, CD3OD): d=174.3, 137.8, 130.2, 128.6, 127.0, 123.1,
122.5, 120.7, 112.5, 107.5, 38.4, 36.6 ppm; IR: n˜ =3389, 3204, 2333, 1752,
1696, 1615, 1529, 1432, 1401 cmꢀ1
; HRMS (ESI): m/z: calcd for
C22H19N4O2 [M+H]+: 371.1503; found 371.1504.
The ligand 3c was further purified by HPLC (ODS column, 0.05% TFA
in CH3CN/0.05% TFA in H2O, 40:60 for 20 min, then 100:0) for meas-
urements of binding properties. The concentration of the stock solution
of 3c was determined by comparing the integration values of the
1H NMR signals with those of maleimide as the internal standard.
1H NMR (400 MHz, CD3OD): d=1.80 (4H, br), 2.05 (2H, m), 2.12 (2H,
m), 2.99–3.20 (10H, m), 3.86 (2H, t, J=6.4 Hz), 7.04 (2H, d, J=8.5 Hz),
7.39 (2H, d, J=7.6 Hz), 7.45–7.49 (4H, m), 7.59 (2H, d, J=7.6 Hz), 7.67
(2H, br), 7.78 ppm (2H, s); 13C NMR (125 MHz, CD3OD): d=172.3,
137.6, 135.1, 134.4, 131.7, 129.7, 129.0, 128.8, 128.6, 127.7, 127.5, 48.2,
48.2, 46.8, 45.8, 37.8, 36.2, 26.9, 25.3, 24.2, 24.2 ppm; HRMS (ESI): m/z:
calcd for C34H39N4O2 [M+H]+: 535.3068; found 535.3066.
Compound 3a: 1H NMR (400 MHz, CD3OD): d=1.83 (4H, brs), 2.11
(4H, m), 3.07–3.12 (10H, m), 3.79 (2H, t, J=6.4 Hz), 6.60 (2H, dd, J=
7.3, 7.9 Hz), 6.83 (2H, d, J=7.9 Hz), 6.97 (2H, dd, J=7.3, 7.9 Hz), 7.33
(2H, d, J=7.9 Hz), 7.75 ppm (2H, s); 13C NMR (125 MHz, CD3OD): d=
174.1, 137.7, 130.1, 128.6, 127.0, 123.0, 122.5, 120.7, 112.4, 107.5, 48.2,
48.1, 46.8, 45.8, 37.8, 35.8, 27.0, 25.3, 24.2, 24.2 ppm; IR: n˜ =3385, 2984,
2817, 2323, 1754, 1689, 1630, 1528, 1454, 1409, 1362 cmꢀ1; HRMS (ESI):
m/z: calcd for C30H37N6O2 [M+H]+: 513.2973; found 513.2952.
Acknowledgements
This work was supported by the Japan Society for the Promotion of Sci-
ence (JSPS) with a Grant-in-Aid of Scientific Research (S), and CREST
from the Japan Science and Technology Agency. We are grateful to Dr.
Yasuo Asami at TA Instruments for the ITC measurements.
[1] A recent review: M. A. Fuertes, V. Cepeda, C. Alonso, J. M. Perez,
Chem. Rev. 2006, 106, 2045–2064.
[2] G. Wang, K. M. Vasquez, Front. Biosci. 2007, 12, 4424–4438.
[3] A. Rich, S. Zhang, Nat. Rev. Genet. 2003, 4, 566–572.
[4] C. Mao, W. Sun, Z. Shen, N. C. Seeman, Nature 1999, 397, 144–146.
[5] R. Tashiro, H. Sugiyama, J. Am. Chem. Soc. 2005, 127, 2094–2097.
[6] A. DꢁUrso, A. Mammana, M. Balaz, A. E. Holmes, N. Berova, R.
Lauceri, R. Purrello, J. Am. Chem. Soc. 2009, 131, 2046–2047.
[7] Y. Zhang, H. Yu, J. Ren, X. Qu, Biophys. J. 2006, 90, 3203–3207.
[8] X. G. Qu, J. O. Trent, I. Fokt, W. Priebe, J. B. Chaires, Proc. Natl.
Acad. Sci. USA 2000, 97, 12032–12037.
Compound 3b: 1H NMR (400 MHz, CD3OD): d=1.83 (4H, brs), 2.11
(4H, brs), 3.03–3.12 (8H, m), 3.78 (2H, t, J=6.4), 7.34–7.44 ppm (10H,
m); 13C NMR (125 MHz, CD3OD): d=172.2, 137.9, 131.0, 130.9, 130.1,
129.5, 48.2, 48.1, 46.7, 45.8, 37.8, 36.1, 26.8, 25.3, 24.2, 24.2 ppm; IR: n˜ =
[9] Y. Xu, Y. X. Zhang, H. Sugiyama, T. Umano, H. Osuga, K. Tanaka,
J. Am. Chem. Soc. 2004, 126, 6566–6567.
11998
www.chemeurj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 11993 – 11999

Spermine–Bisaryl Induces B- to Z-DNA Transition
FULL PAPER
[10] G. Manzini, L. E. Xodo, F. Quadrifolglio, J. H. van Boom, G. A. van
der Marel, J. Biomol. Struct. Dyn. 1987, 5, 651–662.
[11] F.-M. Chen, Nucleic Acids Res. 1988, 16, 2269–2281.
[12] D. Bancroft, L. D. Williams, A. Rich, M. Egli, Biochemistry 1994, 33,
1073–1086.
[20] J. M. Ferreira, R. D. Sheardy, Biophys. J. 2006, 91, 3383–3389.
[21] P. S. Ho, G. J. Quigley, R. Tilton, A. Rich, J. Phys. Chem. 1988, 92,
939–945.
[22] T. F. Kagawa, D. Stoddard, G. W. Zhou, P. S. Ho, Biochemistry 1989,
28, 6642–6651.
[13] H. Ohishi, M. Odoko, K. Grzeskowiak, Y. Hiyama, K. Tsukamoto,
N. Maezaki, T. Ishida, T. Tanaka, N. Okabe, K. Fukuyama, D. Y.
Zhou, K. Nakatani, Biochem. Biophys. Res. Commun. 2008, 366,
275–280.
[14] A. Rodger, S. Taylor, G. Adlam, I. S. Blagbrough, I. S. Haworth,
Biopolymers 1994, 34, 1583–1593.
[15] A. J. Geall, R. J. Taylor, M. E. Earll, M. A. W. Eaton, I. S. Blag-
brough, Chem. Commun. 1998, 1403–1404.
[16] N. Shimada, A. Kano, A. Maruyama, Adv. Funct. Mater. 2009, 19,
3590–3595.
[23] T. F. Kagawa, M. L. Howell, K. Tseng, P. S. Ho, Nucleic Acids Res.
1993, 21, 5978–5986.
[24] R. S. Preisler, H. H. Chen, M. F. Colombo, Y. Choe, B. J. Short, Jr.,
D. C. Rau, Biochemistry 1995, 34, 14400–14407.
[25] S. Nakano, L. Wu, H. Oka, H. T. Karimata, T. Kirihata, Y. Sato, S.
Fujii, H. Sakai, M. Kuwahara, H. Sawai, N. Sugimoto, Mol. BioSyst.
2008, 4, 579–588.
[26] P. L. Gilbert, D. E. Graves, M. Britt, J. B. Chaires, Biochemistry
1991, 30, 10931–10937.
[27] P. A. Mirau, D. R. Kearns, Nucleic Acids Res. 1983, 11, 1931–1941.
[28] M. Britt, F. Zunino, J. B. Chaires, Mol. Pharmacol. 1986, 29, 74–80.
[29] C.-W. Chen, R. H. Knop, J. S. Cohen, Biochemistry 1983, 22, 5468–
5471.
[17] H. Sugiyama, K. Kawai, A. Matsunaga, K. Fujimoto, I. Saito, H.
Robinson, A. H. Wang, Nucleic Acids Res. 1996, 24, 1272–1278.
[18] S. Ikuta, R. Chattopadhyaya, H. Ito, R. E. Dickerson, D. R. Kearns,
Biochemistry 1986, 25, 4840–4849.
Received: April 14, 2010
[19] M. M. Patel, T. J. Anchordoquya, Biophys. Chem. 2006, 122, 5–15.
Published online: September 8, 2010
Chem. Eur. J. 2010, 16, 11993 – 11999
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
11999