SsDNA templated assembly of oligonucleotides and bivalent naphthalene guests

Article abstract of DOI:10.1039/b924631b

The hybridization of oligothymine templates with oligomeric adenine and naphthalene diaminotriazine guests has been studied by UV-vis and circular dichroism (CD) spectroscopy. First, the length of the oligothymine template and the oligoadenine guest has b

Full text of DOI:10.1039/b924631b

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/softmatter | Soft Matter
ssDNA templated assembly of oligonucleotides and bivalent naphthalene
guests
Pim G. A. Janssen, Niels J. M. Brankaert, Xavier Vila and Albertus P. H. J. Schenning*
Received 24th November 2009, Accepted 5th January 2010
First published as an Advance Article on the web 2nd February 2010
DOI: 10.1039/b924631b
The hybridization of oligothymine templates with oligomeric adenine and naphthalene diaminotriazine
guests has been studied by UV-vis and circular dichroism (CD) spectroscopy. First, the length of the
oligothymine template and the oligoadenine guest has been systematically varied. The studies reveal
that increasing the length of multivalent guests results in a stabilization of the templated assemblies.
Hybridization studies also show that guest–guest interactions between the multivalent guests
additionally enhance the stability. Based on these results, a bivalent naphthalene diaminotriazine guest
has been synthesized, showing a higher stability than the monovalent naphthalene diaminotriazine
guest; the host–guest interaction energy is higher while the guest–guest interaction remains the same.
Introduction
great detail. Hence, the hybridization of two single strands under
12
given conditions can be predicted accurately. In an elegant
DNA is a promising building block to construct functional 2D
and 3D nanosized objects via a ‘bottom-up’ approach since it is
easily obtained in a monodisperse form and can be functional-
study, the hybridization of oligomeric strands, dAm (Scheme 1)
13,14
and deoxyuracil nucleotides of equal
14
and unequal length
has been evaluated with a zipper model to get information on the
intra-strand base-stacking interaction. The interaction between
the bases within one strand makes sure that binding of a whole
strand to the complementary strand is favored. In another study,
the binding of a 9-methyl-6-methylaminoadenine monomeric
guest to poly(deoxyuracil) nucleotide (poly(dU)) has been
analyzed with an Ising model for guest binding to infinite-sized
templates where next to the interaction between the guests and
the template a stacking interaction between guests has been taken
1–3
4
ized. DNA has been used as a supramolecular cross-linker
5
and as a scaffold to organize particles, proteins or small
5
6
molecules. In previous studies, we displayed the use of an oli-
gothymine (Tn, Scheme 1) template for the assembly of
complementary diaminotriazine equipped guest molecules (see
6
a–e
for example G1, Scheme 1).
Detailed characterization of the
DNA templated assembly showed that the hybrid structures
have a right-handed helical arrangement and are stabilized by p–
p and hydrogen (H) bonding interactions. The efficiency of this
templated assembly depends on the template–guest and guest–
guest interaction and can be described by a templated assembly
model based on a one-dimensional Ising model. For the exploi-
tation of these ssDNA templated assembly building blocks it is
essential to have high host–guest and guest–guest interactions to
obtain only completely filled templates.
11
into account. These studies reported that the assembly is
cooperative, i.e., a considerable stabilizing guest–guest interac-
tion energy of 3 z ꢀ5kT
p
has been determined between the
adjacent monomeric 9-methyl-6-methylaminoadenines guests on
11
the template. Similarly, the binding of dAm oligomeric guests to
poly(dU) has been analyzed with a zipper model where next to
the host–guest interaction, an intra-strand base-stacking inter-
13
For DNA hybridization, it is known that the transition
action was taken into account. Furthermore, the templated
temperature T
p
(the temperature at which half of the double
15
polymerization of oligomeric peptide nucleic acid and mono-
strands are formed) and the sharpness of the transition highly
depend on the length and base sequence of both strands. Double-
strand formation occurs via a nucleation mechanism where the
first 3 or 4 base pairs formed are energetically unfavorable, after
16
meric nucleotide guests to a larger ssDNA template has been
performed.
Here, we report on the influence of multivalent guests on the
stability of DNA templated assembly by analyzing the hybrid-
ization of oligoadenine (dAm, Scheme 1) with Tn. In this study
the Tn is regarded as a template while dAm is a multivalent guest.
The length of both the oligothymine template (Tn with n ¼ 5, 10,
7,8
which favorable elongation takes place. Hybridization of DNA
is therefore basically an all-or-none process where the elongation
7–10
after the formation of a nucleus behaves like a zipper.
general, longer strands have a higher hybridization transition
temperature, T , and a sharper transition because of higher
binding enthalpy.
In
2
2
0, 40) and the oligoadenine guest (dAm, with m ¼ 1, 2, 4, 5, 8, 10,
p
0, 40) is systematically varied to investigate the influence of the
11
number of guest binding sites on the stability of the formed
The influence of chain ends, nearest neighbor interactions,
salts, and concentration on the hybridization has been studied in
templated assemblies. In contrast with previous studies on the
8,11
hybridization of poly(dU) with dAm (Scheme 1),
ization of Tn with dAm is analyzed with the Ising model for
the hybrid-
Laboratory for Macromolecular and Organic Chemistry, Eindhoven
University of Technology, P.O. Box 513, 5600, MB, Eindhoven, The
Netherlands. E-mail: a.p.h.j.schenning@tue.nl; Fax: +31 40 245 1036;
Tel: +31 402472655
guests binding to a finite-sized template to get information on the
17–20
guest–guest interaction energy.
Secondly, we have synthe-
sized a bivalent naphthalene diaminotriazine guest having
1
494 | Soft Matter, 2010, 6, 1494–1502
This journal is ª The Royal Society of Chemistry 2010

View Article Online
Scheme 1 (a) Chemical structures of template oligothymine (Tn) and guest’s oligoadenine (dAm), G1 and G2. (b) Schematic representation of the Tn
templated assembly of G2.
a flexible linker (G2, Schemes 1 and 3) that can bind to Tn and we
have compared the stability of its templated assembly with its
DNA, q, is obtained by subtracting the slope of DUV versus T at
high temperatures, assuming that the degree of hybridization
eventually converges to one (Fig. 1c). The transition tempera-
6
monomeric analogue (G1).
ture, T
p
, is defined as the temperature at which half of the double
) h 0.5. The change in Cotton
strands are formed, i.e., where q(T
p
Results and discussion
effect reveals a transition similar to the UV-vis measurements.
The CD spectrum at low temperature does not fully resemble the
Hybridization of Tn with dAm
22
spectrum of a poly(dA)–poly(T) B-DNA (Fig. 1b), suggesting
a different supramolecular organization. When fitting the curve
of the temperature-dependent degree of template sites bound, to
the model for templated assembly (Fig. 1c), a guest–guest inter-
action of 3 z ꢀ7.3kT at T ¼ 291 K is found. Remarkably, the
First, the binding of monomeric guest dA1 to T40 is analyzed.
Since it is expected that the host–guest interaction for the twofold
H bond will be significantly lower than for the threefold H bond,
11
a 2 mM concentration of dA1 is used. Upon cooling, while
monitoring the UV-vis absorption of a T40–dA1 mixture,
hypochromicity is observed similar to the hybridization of DNA
p
p
guest–guest interaction of dA1 is higher than our previously
6a
studied G1 guest (3 z ꢀ5.1kT ). A titration of dA1 to T40 at
p
2
1
(
Fig. 1a and c). The hybridization process is further investi-
270 K shows that dA1 binds in a 40 : 1 ratio to T40, i.e., one dA1
23
gated by monitoring the absorption intensity at 275 nm as
a function of temperature. The fraction of double-stranded
per thymine (Fig. 1d). The titration curves at higher tempera-
tures show a lower degree of binding (data not shown). By fitting
This journal is ª The Royal Society of Chemistry 2010
Soft Matter, 2010, 6, 1494–1502 | 1495

View Article Online
Fig. 1 (a) UV-vis and (b) CD spectra of a T40–dA1 mixture at 298 and 266 K. [T]Tn ¼ 0.8 mM and [dA1] ¼ 2 mM. (c) Temperature-dependent
measurement of T40–dA1 at l ¼ 275 nm, [T]Tn ¼ 0.8 mM and [dA1] ¼ 3 mM. (d) Titration at 266 K of dA1 to T40 monitored with the change in
absorption at l ¼ 275 nm, [T]Tn ¼ 0.8 mM and [dA1] ¼ 0–3 mM in a Tris buffer at pH ¼ 7.2 with 10 mM MgCl
2
. The lines in (c) and (d) are fits to the
6
a
templated assembly model.
Fig. 2 (a) UV and (b) CD spectra at 353 and 323 K of T40 and dA40. (c) Degree of hybridization (q) monitored with UV-vis spectroscopy at l ¼ 275 nm
ꢀ
1
with 1 K min of a T40 and dA40 mixture. (d) CD spectra of hybridized T40–dAm mixtures (m ¼ 40 (298 K), 20 (298 K), 10 (298 K), 8 (298 K), 5 (283 K),
4
(283 K), 2 (265 K)). [A]dAm ¼ [T]Tn ¼ 0.5 mM in a 10 mM MgCl
496 | Soft Matter, 2010, 6, 1494–1502
2
Tris buffer at pH ¼ 7.2 in all cases.
This journal is ª The Royal Society of Chemistry 2010
1

View Article Online
2
4
the titration curve of dA1 to T40 at 266 K, a host–guest inter-
action of g z ꢀ6.6kT is estimated (Fig. 1d). This twofold H
bonding host–guest interaction energy is lower than the threefold
H bonding host–guest interaction energy of the diaminotriazine
derivative G1 with T40.
systematically studied using T40 as the template. For each
combination of T40–dAm, the fraction of hybridized strands, q,
is determined by the change in UV absorption at 275 nm, as done
earlier for the T40–dA40 mixture (Fig. 2). For fair comparison,
the base concentration of both components was kept constant at
[T]T40 ¼ [A]dAm ¼ 0.5 mM, similar to the T40–dA40 mixture. As
expected, the stability of the templated assembly increases when
The binding of the multivalent guests is first studied for T40–
24
dA40, using UV-vis and CD spectroscopy. At 353 K, the
T40–dA40 mixture ([T]T40 ¼ [A]dAm ¼ 0.5 mM) only contains
the dAm strand is lengthened as evidenced by the higher transi-
8,11,12
7
single-stranded dA40 and T40 (Fig. 2a and b). Upon cooling,
tion temperature, T
p
(Fig. 3a,d,e).
The Cotton effects for the
DNA hybridization is observed, as indicated by the CD intensity
change (Fig. 2b) and the hypochromicity of the UV-vis absorp-
7
tion (Fig. 2a). The Cotton effect of the double-strand at 323 K is
fully hybridized T40–dAm mixtures (q z 1) have a typical B-
DNA shape, except for the T40–dA2 mixture (q z 0.94)
22
(Fig. 2d). The intensity of the Cotton effect decreases for
similar to what has been reported before for B-DNA formed by
shorter dAm strands, which can be due to more disorder in the
hybridized assemblies. The slope of q versus temperature slightly
increases with the length of dAm due to the higher binding
22
poly(T)–poly(dA). The hybridization process of the T40–dA40
mixture is further investigated by monitoring the absorption
intensity at 275 nm as a function of temperature.
13
enthalpy for longer strands. The fact that dA2 and dA4 guest
The effect of oligomeric length, m, of the adenine guest (dAm,
m ¼ 2, 4, 5, 8, 10, 20, 40) on the hybridrization process is
strands also hybridize with T40 suggests a considerably favorable
11
guest–guest interaction (vide infra).
Fig. 3 Temperature dependence of the degree of hybridization, q, of (a) T40 and dAm mixtures and (b) T40–dA5 and T5–dA40 mixtures at l ¼ 275 nm.
c) CD spectra of hybridized T40–dA5 and T5–dA40 mixtures. (d) T for T40–dAm, T20–dAm and T10–dAm mixtures. (e) T for Tn–dAm mixtures. q
was calculated from the DUV at 275 nm with 1 K min . [A]dAm ¼ [T]T40 ¼ 0.5 mM in 10 mM MgCl Tris buffer at pH 7.2. The lines in (a) and (b) are fits
to the model for templated assembly. For dA20 and dA40 binding to T40 the approximations in the model are not valid and hence fits are not possible.
(
p
p
ꢀ
1
2
6
a
This journal is ª The Royal Society of Chemistry 2010
Soft Matter, 2010, 6, 1494–1502 | 1497

View Article Online
A similar study as for T40–dAm (m ¼ 2, 4, 5, 8, 10, 20, 40)
which the slope is inversely proportional to the enthalpy of
25
mixtures is also performed for shorter oligothymine templates
binding (Fig. 3e). As expected, shorter strands have a lower
enthalpy of binding, implying that the Tp of shorter strands
depends more on concentration. For instance, for T40–dA40 and
T40–dA20, the binding enthalpy is so high that the transition
temperature hardly changes in this concentration range.
24
T20 and T10 (Fig. 3d). Since the template strand Tn is shorter,
the transition temperatures are lower than found for T40. For
p p
example, when comparing the T of T10–dA5 with the T of T40–
dA5, the latter is 22 K higher.
To see whether the interaction between adjacently bound
thymine strands is comparable to adenine strands, a T5 guest is
also hybridized with dA40 template strands and compared to the
T40–dA5 hybridization (Fig. 3b and c). Remarkably, the differ-
To determine the guest–guest interaction, the templated
6a
assembly model previously described is used to fit the
experimental data for guests binding to a finite-sized template
6a
(Scheme 2). The approximations in the model require that the
host must have at least four times more binding sites than the
guest (n > 4m). Oligoadenine dAm is considered as a single guest
and Tn as the template to which n/m guest strands can bind. Only
for T5–dA40 this is reversed; T5 is the guest and dA40 is the
template. For the experiments performed, only the guest–guest
ence in stability is huge. The T of the T5–dA40 mixture is 32 K
p
lower than the T of the T40–dA5 mixture. The Cotton effects of
p
both hybridized mixtures are not the same, indicating different
supramolecular organizations (Fig. 3c). Furthermore, the change
in the degree of hybridization, q, as a function of temperature at
the start of the T5–dA40 hybridization has a lesser slope, sug-
gesting a less cooperative transition, possibly due to a weaker
guest–guest interaction. It can be expected that the host–guest
interaction between the adenines and thymines is the same for
both samples. The difference in stability suggests that the inter-
action between two adjacently bound Tn strands is lower than
between two dAm strands (vide infra).
26
interaction energy, 3, can be determined. The temperature-
dependent T40–dAm hybridizations are fitted to the model for
templated assembly (Fig. 3a); the enthalpy of binding increases
for longer dAm guests (Table 1). Also a considerable guest–guest
p
interaction energy, 3, of several kT units is found for dAm
(Table 1). This interaction energy, 3, decreases with the length of
the guest, but even for dA10 an 3 of ꢀ2.3kT is observed which
p
The concentration dependence of the transition temperature,
stabilizes the hybridization significantly. The decrease of 3 with
the size of the guests can be caused by a different structural order
in the double strands as has been observed with CD spectroscopy
(Fig. 2d). This is possibly due to more electronic repulsion
between the phosphates of neighboring strands or more
conformational freedom of the shorter oligomer guest strands.
For T5–dA40, which is less stable than T40–dA5, 3 is consider-
T , is also measured for four concentrations of T40–dAm
p
mixtures (m ¼ 2, 4, 5, 8, 10, 20, 40). Plotting T against the T40
p
concentration (on a logarithmic scale) shows a linear trend of
ably smaller (ꢀ0.9kT
p
versus ꢀ3.0kT
p
, respectively). This indi-
cates that the interaction between noncovalently linked adenines
is stronger than between noncovalently linked thymines when
bound to the template. The favorable guest–guest interaction
indicates that at high q the templates strands are filled without
19
empty binding sites between the bound guests.
A bivalent naphthalene guest derivative
Scheme 2 A schematic drawing of the Ising chain model for templated
assembly of guest oligomers with m binding sites (spheres with lines).
When a guest adsorbs to a template to which n/m guests can bind, this
gives rise to a negative binding free energy, g. It should be noted that g is
not only a result of the H-bond interaction, but all conformational
changes of template and guests upon binding is also governed in g. When
two guests bind on neighboring sites this liberates an additional free
energy, 3; i.e., 3 is the energetic difference between an isolated guest on the
template and a guest on a template bound next to another guest. This
energy difference can be caused by the stacking interaction between
guests, but also by conformational changes in template and guests. 3 is
negative for the cooperative and positive for the anticooperative process.
In the previous section it has been shown that increasing the
number of binding sites of the guest greatly stabilizes the tem-
plated assembly. The same principle to enhance the stability of
Tn templated self-assemblies can be applied to the naphthalene
6
a
guest G1 reported previously. Therefore, two diaminotriazine
equipped naphthalenes are linked together via an ethylene glycol
to make a bivalent analogue of G1 (G2, Schemes 1 and 3). Unlike
for the dAm guests, the linker for G2 is long and flexible and
hence, the two binding sites are not pre-organized to bind to Tn.
This means that in principle G2 can bind to two template strands
Table 1 Summary of the average fitting parameters of dAm binding to T40 and for T5–dA40 as determined from our experimental data
a
Tn–dAm
T40–dA1
T40–dA2
T40–dA4
T40–dA5
T40–dA8
T40–dA10
T5–dA40
a
T
3
p
/K
(in kT
291
283
ꢀ4.6
ꢀ11
309
ꢀ3.2
ꢀ8.2
ꢀ333
313
ꢀ3.0
ꢀ7.8
ꢀ358
324
ꢀ2.5
ꢀ6.7
ꢀ582
329
ꢀ2.3
ꢀ6.3
ꢀ632
281
ꢀ0.9
ꢀ2.5
ꢀ299
a
p
ꢀ
)
ꢀ7.3
1
a
3/kJ mol
Dh(T
ꢀ21.2
ꢀ
1
a
p
)/kJ mol
ꢀ87.3
ꢀ125
a
For dA1 binding to T40 these parameters could not be obtained at the same concentrations. These values are obtained for [T]Tn ¼ 0.8 mM and [dA1] ¼
3
mM.
1
498 | Soft Matter, 2010, 6, 1494–1502
This journal is ª The Royal Society of Chemistry 2010

View Article Online
ꢂ
2
Scheme 3 Synthetic route towards G2. (a) THF/H O (1 : 1), KOH, TsCl, 2 h at rt. (b) DMF, KOH, 24 h at 80 C. (c) 2-Methoxyethanol, KOH,
dicyandiamide, 8 h at reflux.
as observed for thymine appended bolaamphiphiles binding to
6
G2 is synthesized via a Williamson etherification of 6-cyano-2-
naphthol (3) with dodecaethylene glycol di-p-tosyl ether (2) to
e,27
dAm.
However, because the naphthalenes have a flexible,
hydrophilic linker which can fold back, an effective concentra-
tion Ceff is sensed when one of the naphthalenes is bound to Tn.
The value of this Ceff is the highest at a single p-stacking distance
yield
4
which subsequently reacts with dicyandiamide
(Scheme 3).
Temperature-dependent measurements show that G2 is
molecularly dissolved up to at least a concentration of 50 mM.
However, the naphthalene absorption band is red-shifted from
˚
28
of 3.5 A giving a C value of ꢁ30 mM, suggesting that G2 will
eff
most likely be present as a stacked dimer and bind to two
consecutive binding sites on one T40 strand (Scheme 2).
l
max ¼ 310 nm for G1 to 314 nm for G2 (Fig. 4a). Furthermore,
Fig. 4 (a) Normalized UV-vis spectra of G1 and G2 in water. (b) Relative fluorescence intensity of G1 and G2 in water, lex ¼ 340 nm.
Fig. 5 (a) CD spectra at 323 and 263 K. (b) Cooling curve of T40–G2 mixture (40 : 1 ratio, i.e., two base equivalents) monitored by the CD intensity at
t
l ¼ 270 nm. [G2] ¼ 37.5 mM and [T]T40 ¼ 37.5 mM. (c) T
e
(inverse scale) as a function of the guest concentration [G] (logarithmic scale) for G1 and G2.
(
d) Titration of G2 to T40 at 268 K monitored by the CD intensity at l ¼ 270 nm. [G2] ¼ 0–94 mM and [T]T40 ¼ 50 mM. The lines in (b) and (d) are fits to
6a
the templated assembly model.
This journal is ª The Royal Society of Chemistry 2010
Soft Matter, 2010, 6, 1494–1502 | 1499

View Article Online
the fluorescence of G2 is quenched and red-shifted (Fig. 4b).
These spectral shifts suggest that the two naphthalenes of G2
interact as expected from the high Ceff.
Sample preparation
All samples were prepared by adding a Tn and dAm from
a concentrated solution in MilliQ water to a Tris (tris(hydroxy-
methyl)aminomethane) buffer at pH ¼ 7.2 containing 10 mM
When G2 is mixed with T40 (40 : 1 ratio, i.e., two base
equivalents) a similar Cotton effect in shape and relative intensity
6
Fig. 5a) as earlier found for G1 is observed. The concentration
MgCl
2
. G2 samples were prepared by adding ssDNA in ultrapure
(
ꢂ
water to solid guest and subsequent sonification at 85 C for at
least 15 min. For the titration experiments, two stock solutions
were prepared; one for the starting and one for the end-point of
the titration. For each data point, these were mixed in the
is, however, ten times lower. For a similar guest concentration
t
(
e
25 mM), the apparent elongation temperature, T , for templated
assembly is 299 K for G2, and 263 K for G1 (Fig. 5c), showing
that the T40–G2 complex is more stable than the T40–G1
complex. Similar to dAm binding to Tn, increasing the number of
guest binding sites for the naphthalene guest derivative results in
a more stable templated assembly. Titrating G2 to T40 reveals
a 20 : 1 binding stoichiometry, i.e., one naphthalene per thymine
ꢂ
appropriate ratio and heated to 70 C. The typical equilibration
time before measuring at low or high temperature was 5 min.
Materials
(
Fig. 5d).
Tn and dAm were supplied freeze-dried and HPLC purified by
0
The UV and CD data obtained for G2 and T40 are analyzed
MWG Biotech AG. 2 -Deoxyadenosine dA1 was purchased from
with the model for templated assembly in which G2 is considered
as one guest and T40 can accommodate 20 G2 guests (Fig. 5b).
By fitting the melting curves, a guest–guest interaction energy of
Sigma-Aldrich-Fluka as the mono-hydrate. Dodecaethylene
glycol 1 was supplied by Polypure. All other chemicals were
commercially available and were used without further purifica-
tion.
3
z ꢀ5.3kT
5.1kT ).
p
is determined, which is comparable to G1 (3 z
This similar guest–guest interaction energy
6
a
ꢀ
p
indicates that G2 indeed binds to two consecutive template
binding sites. The host–guest interaction energy (g z ꢀ15kT) is
determined through fitting of the titration experiment (Fig. 5d).
Compared to G1 (g z ꢀ10kT) the host–guest interaction is
significantly improved. It should be noted that the host–guest
interaction is the interaction between a bivalent guest and the
template and that the conformational changes of the dimer upon
binding to the template will appear in the host–guest interaction,
g (Scheme 2).
Dodecaethylene glycol di-p-tosyl ether, 2
ꢂ
Under an argon atmosphere at 0 C, dodecaethylene glycol 1
(
2.5 g, 4.58 mmol) was dissolved in 10 ml THF after which 2 g
KOH dissolved in 10 ml of water was added. After 2 h of stirring
at room temperature, 30 ml of 1 N HCl were added, followed by
an extraction with chloroform. The organic layers were collected,
dried with MgSO
tography (silica, CHCl
as a colorless oil (3.7 g, 95% yield). H NMR d (200 MHz,
CDCl ); 2.43 (s, 6H, CH ), 3.54–3.80 (m, 44H, OCH ), 4.14 (t,
H, OCH
), 7.32 (d, 4H, ArH, J ¼ 8.2 Hz), 7.78 (d, 4H, ArH,
J ¼ 8.2).
4
and further dried in vacuo. Column chroma-
3
: MeOH from 100 : 0 to 98 : 2) afforded 2
1
3
3
2
4
2
Conclusion
0
The high guest–guest interaction of dA1 makes 2 -deoxynucleo-
Dodecaethylene glycol di-6-cyano-2-naphthyl ether, 4
sides an interesting building block for ssDNA templated
assembly. The hybridization of oligothymine with oligoadenine
shows that the interaction between two adjacently bound strands
stabilizes the DNA hybridization in such a way that even strands
smaller than 4 bases can bind. The double-strand stability
decreases rapidly for smaller strands. Furthermore, the DNA
strand concentration for strand lengths longer than 10 hardly
ꢂ
Under an argon atmosphere at 0 C dodecaethylene glycol p-
tosyl ether 2 (3.77 g, 4.33 mmol) in 5 ml dry DMF was added
dropwise to a solution of 6-cyano-2-naphthol 3 (2.1 g, 12.4
mmol) and KOH (800 mg, 14.3 mmol) in dry DMF (20 ml). After
ꢂ
1
5 min the reaction mixture was heated to 80 C for 24 h. The
reaction mixture was added to 40 ml 1 N HCl and extracted with
chloroform. The collected organic phase was washed neutral
p
affects the transition temperature, T . Increasing the host–guest
interaction seems to be easiest by linking two single guests. In
case of the nonnatural bivalent guest, the concentration at which
binding occurs is ten times lower because of the higher host–guest
interaction energy revealing that the use of multivalent guest
molecules is an appealing approach to enhance the efficiency in
templated assembly.
with water and dried with MgSO . Dodecaethylene glycol di-6-
4
cyano-2-naphthyl ether 4 (1.78 g, 47% yield) was obtained pure
as a white solid after column chromatography (active neutral
1
alumina oxide, CHCl : MeOH ¼ from 100 : 0 to 98 : 2). H
3
3 2
NMR d (400 MHz, CDCl ); 3.54–3.73 (m, 40H, OCH ), 3.93 (t,
4
H, OCH
2
CH
2
ONaph), 4.27 (t, 4H, NaphOCH
¼ 2.4, J
d, 2H, NaphH, J ¼ 8.8), 7.76 (d, 2H, NaphH, J ¼ 8.8 Hz), 8.11
s, 2H, NaphH). MALDI-TOF MS; (M ¼ 848.40) m/z ¼ 871.44
2
), 7.15 (d, 2H,
NaphH, J ¼ 2.4 Hz), 7.27 (dd, 2NaphH, J
1
2
¼ 8.8), 7.54
(
(
Experimental
+
+
[M + Na ] .
General methods
Dodecaethylene glycol di-6-diaminotriazine-2-naphthyl ether,
G2
CD and UV-vis spectra were recorded on a JASCO 815 equipped
with a Peltier temperature controller, PFD-425S. All spectra
ꢀ1
were recorded by cooling 1 K min from a temperature at which
all strands are single strands.
Under an argon atmosphere, 4 (1.73 g, 2.05 mmol) and dicyan-
diamide (190 mg, 2.26 mmol, erroneously only 1.1 equivalent of
1
500 | Soft Matter, 2010, 6, 1494–1502
This journal is ª The Royal Society of Chemistry 2010

View Article Online
dicyandiamide was used) were dissolved in 10 ml 2-methoxy-
ethanol. After adding KOH (400 mg, 7.18 mmol) the reaction
mixture was stirred overnight at reflux. The mixture was added to
Greef, M. R. J. Vos, P. H. H. Bomans, N. A. J. M. Sommerdijk,
P. C. M. Christianen, Ph. Lecl ꢀe re, R. Lazzaroni, P. van der Schoot,
E. W. Meijer and A. P. H. J. Schenning, J. Am. Chem. Soc., 2009,
1
31, 1222; (b) P. G. A. Janssen, J. Vandenbergh, J. L. J. van
1
N HCl (20 ml) and extracted with chloroform. The organic
layers were combined, washed with water and dried with MgSO
Extensive column chromatography (silica and active neutral
alumina oxide, both CHCl
Dongen, E. W. Meijer and A. P. J. H. Schenning, J. Am. Chem.
Soc., 2007, 129, 6078; (c) P. G. A. Janssen, J. L. J. van Dongen,
E. W. Meijer and A. P. H. J. Schenning, Chem.–Eur. J., 2009, 15,
4
.
3
52; (d) M. Surin, P. G. A. Janssen, R. Lazzaroni, E. W. Meijer
3
: MeOH ¼ from 100 : 0 to 90 : 10)
and A. P. H. J. Schenning, Adv. Mater., 2009, 21, 1126; (e)
R. Iwaura, F. J. M. Hoeben, M. Masuda, A. P. H. J. Schenning,
E. W. Meijer and T. Shimizu, J. Am. Chem. Soc., 2006, 128, 13298;
and preparative reversed phase HPLC (C18 column, water–
acetonitrile) yielded G2 (210 mg, 10% yield) as a white powder.
1
H NMR d (400 MHz, DMSO); 3.32–3.54 (m, 36H, OCH
(
f) B. A. Armitage, Mol. Supramol. Photochem., 2006, 14, 255; (g)
2
), 3.59
R. Iwaura, M. Ohnishi-Kameyama and T. Iizawa, Chem.–Eur. J.,
2009, 15, 3729; (h) M. Banchelli, D. Berti and P. Baglioni, Angew.
Chem., Int. Ed., 2007, 46, 3070; (i) H. A. Becerrril and
A. T. Woolley, Chem. Soc. Rev., 2009, 38, 329.
W. Saenger, Principles of Nucleic Acid Structure, Springer-Verlag,
New York, 1984.
8 D. P o€ rschke, Mol. Biol., Biochem. Biophys., 1977, 24, 191.
(
t, 4H, OCH ), 3.80 (t, 4H, OCH CH Naph), 4.22 (t, 4H,
2
2
2
NaphOCH ), 6.79 (s, 4H, NH ) 7.19 (dd, 1H, NaphH, J ¼ 2.0
2
2
1
and J ¼ 8.8 Hz), 7.35 (d, 1H, NaphH, J ¼ 2.4 Hz), 7.82 (d, 1H,
2
7
NaphH, J ¼ 8.8 Hz), 7.91 (d, 1H, NaphH, J ¼ 8.8 Hz), 8.27 (d,
13
1
H, NaphH, J ¼ 8.8 Hz), 8.71 (s, 1H, NaphH). C NMR d (300
9
For the description of DNA hybridization usually for simplicity
a two-state model is used. See references 7 and 8, and for example:
MHz, CDCl
3
); 0.6, 39.3, 39.5, 39.7, 40.0, 40.2, 40.4, 40.6, 67.8,
6
1
7
1
3
+
9.3, 70.2, 70.3, 70.4, 107.1, 119.6, 125.7, 126.9, 128.2, 131.0,
(
a) G. Bonnet, O. Krichevsky and A. Libchaber, Proc. Natl. Acad.
ꢀ1
32.6, 136.4, 139.7, 158.0, 167.6, 170.4. IR (KBr): n/cm ¼ 735,
64, 813, 827, 862, 913, 954, 976, 1061, 1092, 1178, 1219, 1258,
371, 1387, 1402, 1441, 1480, 1502, 1532, 1581, 1623, 1668, 2867,
166, 3344. MALDI-TOF MS; (M ¼ 1016.50) m/z ¼ 1017.43 [M
Sci. U. S. A., 1998, 95, 8602; (b) J. Kim, S. Doose, H. Neuweiler
and M. Sauer, Nucleic Acids Res., 2006, 34, 2516; (c) M. I. Wallace,
L. Ying, S. Balasubramanian and D. Klenerman, Proc. Natl. Acad.
Sci. U. S. A., 2001, 98, 5584.
0 For DNA hybridization it is often necessary to include an
intermediate state. See for example: (a) J. Jung, R. Ihly,
E. Scott, M. Yu and A. van Orden, J. Phys. Chem. B, 2008,
1
+
+
H ] .
1
2
12, 127; (b) J. Jung and A. van Orden, J. Am. Chem. Soc.,
006, 128, 1240.
Acknowledgements
1
1
1 G. W. Hoffmann and D. P o€ rschke, Biopolymers, 1973, 12, 1611.
2 For hybridization studies of two single strands that bind in a 1 : 1
fashion and predictions of the thermodynamic parameters, see for
example: (a) J. Santalucia, Jr, Proc. Natl. Acad. Sci. U. S. A., 1998,
95, 1460; (b) H. T. Allawi and J. SantaLucia, Jr, Biochemistry, 1997,
The authors acknowledge Xianwen Lou for the MALDI-TOF
spectra, Paul van der Schoot and Bert Meijer for scientific
discussions, Koen Pieterse for the artwork and the EURYI
scheme for financial support.
3
3 D. P o€ rschke, Biopolymers, 1971, 10, 1989.
6, 10581.
1
1
4 Y. Blagoi, V. Zozulya, S. Egupov, V. Onishchenko and
G. Gladchenko, Biopolymers, 2007, 86, 32.
1
5 R. E. Kleiner, Y. Brudno, M. E. Birnbaum and D. R. Liu, J. Am.
Chem. Soc., 2008, 130, 4646.
Notes and references
1
2
N. C. Seeman, Nature, 2003, 421, 427.
16 (a) T. Inoue and L. E. Orgel, Science, 1983, 219, 859; (b) X. Li,
Z.-Y. J. Zhan, R. Knipe and D. G. Lynn, J. Am. Chem. Soc.,
2002, 124, 746.
17 In this study, q has been determined with intervals of 0.1 K, which is
needed for determining the cooperativity accurately at the start of the
hybridization process.
18 Although the nearest neighbor interactions in DNA hybridization
have been studied in great detail, the effect of the interaction
between two strands that bind to two neighboring sites on
a template strand has thus far not been described.
Examples of 2D self-assembly of DNA: (a) N. C. Seeman, H. Wang,
X. Yang, F. Liu, C. Mao, W. Sun, L. Wenzler, Z. Shen, R. Sha,
H. Yan, M. H. Wong, P. Sa-Ardyen, B. Liu, H. Qiu, X. Li,
J. Qi, S. M. Du, Y. Zhang, J. E. Mueller, T.-J. Fu, Y. Wang and
J. Chen, Nanotechnology, 1998, 9, 257; (b) E. Winfree, F. Liu,
L. A. Wenzler and N. C. Seeman, Nature, 1998, 394, 539;
(
c) P. W. K. Rothemund, Nature, 2006, 440, 297.
3
4
Examples of 3D self-assembly of DNA: (a) J. Chen and N. C. Seeman,
Nature, 1991, 350, 631; (b) Y. Zhang and N. C. Seeman, J. Am. Chem.
Soc., 1994, 116, 1661; (c) W. M. Shih, J. D. Quispe and G. F. Joyce,
Nature, 2004, 427, 618; (d) R. P. Goodman, R. M. Berry and
A. J. Turberfield, Chem. Commun., 2004, 1372.
19 For a complete description of the model, see reference 6a.
20 The study is performed in a Tris buffer at pH ¼ 7.2 containing 10 mM
2
MgCl ; conditions at which triple helices are not formed, see:
Examples where DNA is used to cross-link supramolecular objects:
(
S. Arnott, D. W. L. Hukins, S. D. Dover, W. Fuller and
A. R. Hodgson, J. Mol. Biol., 1973, 81, 107.
a) T. Maruyama, H. Yamamura, M. Hiraki, Y. Kemori,
H. Takata and M. Goto, Colloids Surf., B, 2008, 66, 119; (b)
A. B. Dahlin, M. P. Jonsson and F. Hoeoek, Adv. Mater., 2008, 20,
21 dA1 without the T40 template does not show any transition by
temperature-dependent measurements.
1
436; (c) M. Banchelli, F. Betti, D. Berti, G. Caminati, F. Baldelli
22 (a) D. M. Gray, K. H. Johnson, M. R. Vaughan, P. A. Morris,
J. C. Sutherland and R. L. Ratliff, Biopolymers, 1990, 29, 317; (b)
S. T. Gudibande, S. D. Jayasena and M. J. Behe, Biopolymers,
1988, 27, 1905.
Bombelli, T. Brown, L. M. Wilhelmsson, B. Norden and
P. Baglioni, J. Phys. Chem. B, 2008, 112, 10942.
Examples where DNA is used to organize nanoparticles and proteins:
5
(
a) H. Yan, S. H. Park, G. Finkelstein, J. H. Reif and T. H. LaBean,
23 For each data point, the fraction of occupied binding sites of the T40,
q, at a given temperature is calculated from the temperature-
dependent UV-vis spectroscopy measurement.
24 It should be noted that for dA20 and dA40 binding to T40 or T20,
double strands longer than T40 and T20, respectively, can be formed.
25 L. A. Marky and K. J. Breslauer, Biopolymers, 1987, 26, 1601.
26 We have use the McGhee–von Hippel model to obtain information
on the host–guest and guest–guest interaction, but these attempts
failed. See: J. D. McGhee and P. H. Von Hippel, J. Mol. Biol.,
1974, 86, 469.
Science, 2003, 301, 1882; (b) S. H. Park, P. Yin, Y. Liu, J. H. Reif,
T. H. LaBean and H. Yan, Nano Lett., 2005, 5, 729; (c)
S. Srivastava and N. A. Kotov, Soft Matter, 2009, 5, 1146; (d)
M. E. Leunissen, R. Dreyfis, F. C. Cheong, D. G. Grier, R. Sha,
N. C. Seeman and P. A. Chaikin, Nat. Mater., 2009, 8, 590; (e)
J. Sharma, R. Chhabra, A. Cheng, J. Brownell, Y. Liu and H. Yan,
Science, 2009, 323, 112; (f) S. Y. Park, A. K. R. Lytton-Jean,
B. Lee, S. Weigand, G. C. Schatz and C. Mirkin, Nature, 2008, 451,
553.
Examples of DNA templated assembly: (a) P. G. A. Janssen,
S. Jabbari-Farouji, M. Surin, X. Vila, J. C. Gielen, T. F. A. de
6
27 (a) R. Iwaura, K. Yoshida, M. Masuda, M. Ohnishi-Kameyama,
M. Yoshida and T. Shimizu, Angew. Chem., Int. Ed., 2003, 42,
This journal is ª The Royal Society of Chemistry 2010
Soft Matter, 2010, 6, 1494–1502 | 1501

View Article Online
2
1
009; (b) R. Iwaura, Y. Kikkawa, M. Ohnishi-Kameyama and
p-stacking distance, hr
e
i is the squared average end-to-end distance
2
2
˚
T. Shimizu, Org. Biomol. Chem., 2007, 5, 3450; (c) R. Iwaura,
M. Ohnishi-Kameyama and T. Shimizu, Chem. Commun., 2008, 5770.
8 Ceff is calculated via: Ceff ¼ (1.5phr )
an ethylene glycol spacer with 12 units, where r
hr
e
i ¼ C
n
nb , with b ¼ 1.46 A, n ¼ 3 ꢃ ethylene glycol units C ¼
n
ꢀ
1
4.1. Note that Ceff is in mol l . See: S. P. Powers, I. Foo, D. Pinon,
U. G. Klueppelberg, J. F. Hedstrom and L. J. Miller, Biochemistry,
1991, 30, 676.
2
ꢀ3/2
ꢀ1
2
e
2
2
e
i ) (NAV
exp(ꢀ3r
/2hr
e
i ) for
˚
e
¼ 3.5 A is the
1
502 | Soft Matter, 2010, 6, 1494–1502
This journal is ª The Royal Society of Chemistry 2010