Enhancing the melting properties of small molecule-DNA hybrids through designed hydrophobic interactions: An experimental-computational study

Article abstract of DOI:10.1021/ja300322a

Detailed experimental and computational studies revealed the important role that hydrophobic interactions play in the aqueous assembly of rigid small molecule-DNA hybrid (rSMDH) building blocks into nanoscale cage and face-to-face (ff) dimeric structures.

Full text of DOI:10.1021/ja300322a

Article
pubs.acs.org/JACS
Enhancing the Melting Properties of Small Molecule-DNA Hybrids
through Designed Hydrophobic Interactions: An Experimental-
Computational Study
Ibrahim Eryazici, Ilyas Yildirim, George C. Schatz,* and SonBinh T. Nguyen*
Department of Chemistry, Northwestern University, 2145 Sheridan Road, Evanston, Illinois 60208, United States
S
* Supporting Information
ABSTRACT: Detailed experimental and computational studies
revealed the important role that hydrophobic interactions play
in the aqueous assembly of rigid small molecule-DNA hybrid
(rSMDH) building blocks into nanoscale cage and face-to-face
(ff) dimeric structures. In aqueous environments, the hydro-
phobic surfaces of the organic cores in these nanostructures are
minimized by interactions with the core in another rSMDHs,
with the bases in the attached DNA strands, and/or with the
base pairs in the final assembled structures. In the case that the
hydrophobic surfaces of the cores could not be properly isolated
in the assembly process, an ill-defined network results instead of
dimers, even at low concentration of DNA. In contrast, if ff
dimers can be formed with good minimization of the exposed
hydrophobic surfaces of the cores, they are highly stable structures with enhanced melting temperatures and cooperative melting
behavior.
anthraquinones with DNA bases or base pairs has been utilized
to stabilize small duplex DNA-detection probes to improve
their target affinity and specificity.50,51 In a related report, small
DNA hairpins have been shown to exhibit enhanced stability
when they are end-capped with rigid organic molecules.52
Given these precedents, it is surprising that only a few studies
to date have examined the roles that hydrophobic interactions
play in the assembly of organic-DNA building blocks into larger
structures. Notably, perylene diimide (PDI)-DNA hybrids have
been found to form both hairpin dimers53 based on double-
helix DNA assembly as well as larger supramolecular oligomers
involving hydrophobic interactions of the PDI cores with each
other via hydrophobic stacking.16,18
Herein, we present a comprehensive experimental and
computational study to understand the roles that rigid,
hydrophobic organic cores play in the assembly of rigid small
molecule-DNA hybrids (rSMDHs), which are molecules
comprosed of several ss-DNAs attached to a rigid organic
core, into discrete nanostructures (Figure 1). We found the
final nanostructures to have been assembled in aqueous
environments in a way that minimizes the exposure of the
hydrophobic surface of the organic core in rSMDH, through
interactions with the core in another rSMDH, with the bases in
the attached DNA strands, and/or with the base pairs in the
final assembled structures. To evaluate the importance of these
hydrophobic interactions, we prepared two distinct dimer
INTRODUCTION
■
Among natural biopolymers, DNA and RNA possess the
unique ability to self-assemble into a unique double-stranded
helix form in ionic media, based on a combination of H-
bonding of bases, π−π stacking of base pairs, and electrostatic
stabilization of the phosphate backbone.1 The stable DNA-
based helix, with a diameter of ∼2 nm and persistence length of
∼50 nm, has long been used as the sole building block for
engineering 2D2−5 and 3D6−9 nanostructures that are used to
pattern nanoparticles10,11 and biomolecules12−14 on solid
surfaces for electronic and biosensor applications. DNA has
also been linked to rigid15−24 and flexible (dendritic25−27
)
organic molecules, polymers,28−30 metal complexes,31−37 and
nanoparticles38,39 to be used for highly selective DNA
detection40−42 and to direct the assembly of those par-
ticles43−47 and complexes48,49 in a predesigned fashion for
electronic applications.
The assembly of organic-DNA hybrids into nanostructures in
aqueous media has generally been explored using three basic
parameter groups: (1) the number of single strand (ss) DNA
chains that are attached to the organic “core” fragment; (2) the
orientation, geometry, and concentration of these strands; and
(3) the type (Na+, K+, Ca2+, Mg2+, etc.) and concentration of
salts. Surprisingly, the role of the organic core in the assembly
of nanoscale organic-DNA structures has rarely been
investigated even though the core fragment can often be
quite hydrophobic and should have significant interactions with
the bases/base pairs in ionic media. Indeed, the hydrophobic
interaction of rigid organic groups such as stilbenes and
Received: January 12, 2012
Published: March 19, 2012
© 2012 American Chemical Society
7450
dx.doi.org/10.1021/ja300322a | J. Am. Chem. Soc. 2012, 134, 7450−7458

Journal of the American Chemical Society
Article
containing three DNA arms (rSMDH3), each comprising 15-bp
complementary and 6-bp (T6) non-complementary bases, at
low concentrations.21 Similar cage assemblies comprising DNA-
hybrids have also been mentioned by the von Kiedrowski,26
Southern,27and Sawai groups.19 The theses of these previous
papers were primarily focused on the assembly and character-
ization of these cage assemblies as well as their melting
properties; the effects of the cores and non-complementary
spacers in the assembly process were very much ignored. In
light of recent reports on the important roles that hydrophobic
interactions and steric constraints may play in the assembly of
DNA-linked nanostructures,16,18 we became interested in
elucidating the effects of non-complementary Tn (n = 0, 3,
and 6) spacers upon the assembly and stability of cage dimers
possessing large hydrophobic organic cores. As part of this
effort, we improved the stability of the linkage of our organic
core (C) to DNA by adding an extra glycolic moiety
(−CH2CH2O−) to all the termini (see core C in Figure 2)
of our original 1,3,5-tris(p-ethynylphenyl)benzene organic
core.23 We also improved the earlier core synthesis to make
it more conducive to scale-up and purification (see Scheme S1
in the Supporting Information for the synthesis of core C).
The synthesis of symmetric rSMDH3 was achieved in good
yields in two steps: the phosphoramidite core C is first attached
to either the 3′ or 5′ end of single-strand oligonucleotides (15-
to-21-base pair (bp)) grown from the surface of controlled
porosity glass beads (CPG). The second and third arms are
then simultaneously grown from the resulting core C-
functionalized CPG via 3′-normal or 5′-reverse phosphorami-
dite chemistry (Figure 2). The final DMT-protected products
were then cleaved from the solid support and purified by
reverse-phase HPLC, and then DMT is deprotected to yield the
desired rSMDH3's. This synthesis afforded a library of
rSMDH3's, in which three of the same single-strand
oligonucleotides (15 complementary and 0−6 (T0, T3, and
T6) non-complementary bases) are connected to the rigid core
C from either the 3′ or 5′ end (Table 1, entry 1−9). The purity
of all rSMDH3's was ascertained via analytical reverse-phase
HPLC and their length and base compositions were confirmed
via MALDI-ToF mass spectrometry (see Supporting Informa-
tion for more details).
Figure 1. A schematic representation of the assembly of rSMDH3-
XT6-5′ and its complementary rSMDH3-X′T6-5′ into a cage dimer (A)
and the assembly of rSMDH3-XT0-3′ and its complementary rSMDH3-
X′T0-5′ into a ff dimer (B). Gray triangle depicts an organic core; [c]
denotes the total concentration of the ss-DNA arms.
assemblies (cage and face-to-face (ff)) from complementary
rSMDH3's where the solvent-accessible surface area (SASA) of
the hydrophobic cores can be tuned by polythymine spacers or
by controlling the orientation of the three attached ss-DNA
strands (Figure 1). Compared to the cage dimer (Figure 1A),
which melts (i.e., dehybridizes) at 3 °C higher than its DNA
duplex components, the ff assembly (Figure 1B) showed a
dramatic enhancement in melting properties (by 15 °C) due to
interaction of the two rigid, hydrophobic cores with each other
and their “intercalation” into the DNA assembly, as explained
by molecular dynamics (MD) simulations. These results
strongly suggest that hydrophobic interactions can significantly
affect the structures and properties of nanostructures
constructed from organic-DNA hybrids. Thus, hydrophobic
interactions should be considered as a major parameter in the
design of nanoscale assemblies from organic-DNA (or polymer-
DNA) building blocks, together with the number of ss-DNA
surrounding the organic core; the orientation, geometry and
concentration of the strands; and the type and concentration of
the salts in the assembly media.
RESULTS AND DISCUSSIONS
Self-Assembly of Cage Dimers. Combining equimolar
amounts of rSMDH3-XTn-5′ (n = 0, 3, and 6) and its
complement rSMDH3-X′Tn-5′ (n = 0, 3, and 6) (or rSMDH3-
YT3-3′ and its complement rSMDH3-Y′Tn-3′) at low concen-
■
Synthesis of rSMDH3 Building Blocks. We previously
reported the formation of cage dimers (Figure 1A) based on
the hybridization of two rigid small molecule-DNA hybrids
Figure 2. Schematic presentation of the synthesis of rSMDH3-XTn-5′ (Table 1, n = 0, 3, and 6) using 3′- and 5′-phosphoramidite chemistry. Other
sequences listed in Table 1 were synthesized using the same synthetic methodology.
7451
dx.doi.org/10.1021/ja300322a | J. Am. Chem. Soc. 2012, 134, 7450−7458

Journal of the American Chemical Society
Article
Table 1. Single-strand rSMDH3 Oligonucleotides Used in
This Work
entry
short name
orientation
1
2
3
4
5
6
7
8
9
rSMDH3-XT0-5′
rSMDH3-XT3-5′
rSMDH3-XT6-5′
rSMDH3-X′T0-5′
rSMDH3-X′T3-5′
rSMDH3-X′T6-5′
rSMDH3-XT0-3′
rSMDH3-YT3-3′
rSMDH3-Y′T3-3′
(3′-X-5′)3C
(3′-X-T3-5′)3C
(3′-X-T6-5′)3C
(3′-X′-5′)3C
(3′-X′-T3-5′)3C
(3′-X′-T6-5′)3C
(5′-X-3′)3C
(5′-Y-T3-3′)3C
(5′-Y′-T3-3′)3C
sequence X: 3′-ATC CTT ATC AAT ATT-5′
sequence X′: 5′-TAG GAA TAG TTA TAA-3′
sequence Y: 3′-TTA TAA CTA TTC CTA-5′
sequence Y′: 5′-AAT ATT GAT AAG GAT-3′
Figure 3. Non-denaturing PAGE-gel image (6%) of DNA assemblies
with (A) 2 μM total ss-DNA concentration. From left to right: lane 1
= HL5 DNA ladder, lane 2 = control-rSMDH3-T0, lane 3 = cage-T0-
T0, lane 4 = control-rSMDH3-T3, lane 5 = cage-T3-T3, lane 6 =
control-rSMDH3-T6, lane 7 = cage-T6-T6, lane 8 = cage-T3-T3-reverse;
(B) 1 μM total ss-DNA concentration. From left to right: lane 1 =
cage-T6-T6, lane 2 = cage-T3-T3-reverse, lane 3 = cage-T0-T0, lane 4 =
HL5 DNA ladder. Please refer to Table 2 for explanations of the short
names.
trations (2 μM total ss-DNA concentration, in which each arm’s
concentration in the rSMDH3 was calculated separately) were
expected to result in cage structures (Figure 1A, listed in Table
2) as previously reported.19,21,26,27 A low-concentration
Table 2. Double-strand rSMDH3 Assemblies, Including Cage
Dimers and Their Controls, Used in This Work
mobilities of these combinations on non-denaturing PAGE-gel
against those of the [rSMDH3-X′Tn-5′:3X] control groups.
These latter groups provides a better reference point than
commercially available linear DNA ladders (i.e., HyperLadder
V) for sizes and mobilities of three-arm rSMDH3's (Table 2) in
a PAGE-gel study because the DNA arms are linked to one
organic core and the overall assembly can better reflect the
charge/hydrodynamic volume ratio for rSMDH3's. We note
with interest that all control assemblies, control-rSMDH3-Tn (n
= 0, 3, and 6; Table 2, entries 5−6), displayed several bands
around a single dark spot in the gel (Figure 3A, lanes 2, 4, and
6) that can be attributed to the different conformations of the
arms in the assemblies, distinguishable at 4 °C where the
PAGE-gel experiment was carried out.54
a
entry
components
short name
b
b
1
2
3
4
5
6
7
rSMDH3-XT0-5′:rSMDH3-X′T0-5′
rSMDH3-XT3-5′:rSMDH3-X′T3-5′
rSMDH3-XT6-5′:rSMDH3-X′T6-5′
rSMDH3-YT3-3′:rSMDH3-Y′T3-3′
rSMDH3-X′T0-5′:3X
cage-T0-T0
cage-T3-T3
cage-T6-T6
cage-T3-T3-reverse
control-rSMDH3-T0
control-rSMDH3-T3
control-rSMDH3-T6
rSMDH3-X′T3-5′:3X
rSMDH3-X′T6-5′:3X
a
b
Total DNA concentration of 1 and 2 μM. This structure did not
form according to PAGE-gel study (see Figure 3).
experimental range is required for cage dimer formation
because at higher concentrations (>2 μM total ss-DNA
concentration) rSMDH3's tend to form other ill-defined
networks much larger in size than the dimer cages.20,21
Non-denaturing PAGE-gel analysis of the [rSMDH3-XTn-
5′:rSMDH3-X′Tn-5′] and [rSMDH3-YTn-5′:rSMDH3-Y′Tn-5′]
combinations confirmed the formation of the expected cage
Tn-Tn (n = 3 and 6) structures (Figure 3). The three
combinations that we attempted all afforded a main band
(Figure 3A, lanes 5, 7, and 8) that is slightly slower in mobility
compared to the corresponding [rSMDH3-X′Tn-5′:3X] control
groups (Figure 3A, lanes 2, 4, and 6), which were prepared by
hybridizing 3 equiv of single-stranded 3′-X-5′ with rSMDH3-
X′Tn-5′ (n = 0, 3, and 6, see Table 2, entries 5−7). In contrast,
the T0-T0 combinations did not form the intended cages: the
[rSMDH3-XT0-5′:rSMDH3-X′T0-5′] combination yielded a
high-molecular-weight single band that did not move in the
non-denaturing PAGE-gel (Figure 3A, lane 3). Such a high-
molecular-weight material has been characterized by Sleiman
and co-workers using atomic force microscopy (AFM)20 as
being an ill-defined network comprising hundreds of rSMDH3's
and being much bigger than the 500-bp DNA ladder (Figure
3A, lane 1).
We also attempted to assemble cage-T0-T0 at a lower DNA
concentration (1 μM), knowing that this would favor the
formation of discrete cage dimers over ill-defined network.23
Nevertheless, the PAGE-gel experiment again showed exclusive
formation of high-molecular-weight networks that did not
move on the gel (Figure 3B). Given that the other rSMDH3's
with non-complementary T3 and T6 spacers can form cages
(Figure 3A), we suspected that concentration is not the only
factor that dictate the formation of small cages or large
networks. Rather, the presence of the non-complementary
thymine spacers can significantly influence the product
formation. As such, we turn to MD simulations to help us
elucidate the difference in assembly formation between
rSMDH3 with T0 spacer and those with T3 and T6 spacers.
Molecular Dynamics Simulations of Cage Systems.
For small DNA duplexes that were assembled from organic-
DNA hybrids, the interactions between the organic group with
DNA bases and base pairs can exert a significant influence on
the final properties of the assembled organic-DNA duplexes. In
contrast, the influence of the organic group in the assembly
process and the final properties of large DNA-hybrid structures
comprising more than one DNA duplex have not been
evaluated in detail by computational modeling. While computa-
tional tools such as coarse-grain simulations55,56 are available,
they are low-resolution and were mainly used to simulate large
DNA nanostructures to study their melting, mechanical, and
Evidence for cage formation in the [rSMDH3-XTn-
5′:rSMDH3-X′Tn-5′] and [rSMDH3-YTn-5′:rSMDH3-Y′Tn-5′]
combinations is afforded by the close similarities in the
7452
dx.doi.org/10.1021/ja300322a | J. Am. Chem. Soc. 2012, 134, 7450−7458

Journal of the American Chemical Society
Article
conformational properties.55−58 In-depth, atomistic-level com-
putational methods have not been applied to study the
assembly of DNA-hybrids into large structures (>20 nm), so-
called supramolecular DNA assemblies,59 due to their inability
to give meaningful data for time scales that are comparable to
the dynamics of biological polymers. However, MD simulations
with atom-based force fields, such as AMBER60−62 and
CHARMM,63,64 are starting to gain ground through improve-
ments in computational technology.65 Millisecond-long atom-
istic-level MD simulations can now be performed to shed light
on the assembly mechanism of complex systems.66 As such, we
set out to model the assembly of the aforementioned
[rSMDH3-XTn-5′:rSMDH3-X′Tn-5′] assemblies using MD
simulation.
Model systems for cage-Tn-Tn (n = 0, 3, and 6) were created
with the AMBER60−62 simulation packages. The organic core
was parametrized with the AMBER GAFF force field,67 and
restrained electrostatic potential (RESP) charges of the core’s
atoms were derived following the RESP protocol (see
Supporting Information for details).68−71 For DNA duplexes,
amber99 force field,69,72 with revised χ73 and α/γ74 torsional
parameter sets, were used. Initial structures of the cage dimers
were constructed with shortened versions of rSMDH3-XTn-5′
and rSMDH3-X′Tn-5′, each comprising three 10-bp ds-DNA in
B-form conformation that are attached to the organic cores. In
the initial configuration, the cores were positioned in the model
as far away as possible from DNA bases and base pairs (Figure
4A−C). For each assembly, four independent, implicit-
solvent75,76 MD simulations were performed and each
simulation was 100 ns long. At the end of the simulations,
the cores in cage-T3-T3 and cage-T6-T6 (Figure 3E,F) were
observed to be uniformly interacting with non-complementary
thymine bases. The core for cage-T0-T0 was observed to
intercalate into the closest base pairs, which significantly
“deform” the structure (Figure 4D). These results suggest that
the cage assemblies have reorganized to minimize the
hydrophobic surfaces of the cores in aqueous environments
at the detriment of the caged structure.
To qualitatively estimate the hydrophobic surface area of the
core that is exposed to water throughout the simulation (100
ns), we performed a SASA analysis of trimer cores in the cage
assemblies using the AREAIMOL program (ver. 6.2.0) from
CCP4.77 With only the benzyl and acetylenes moieties of the
trimer cores included as hydrophobic moieties in these
calculations, the total SASA of two fully exposed cores in
each cage assembly at the start of the simulations was calculated
to be ∼1620 Å2. Then the final SASA data of the assemblies
were calculated by averaging the last 50 ns of each set of
simulations (Table S6 and Figures S41−S46 in Supporting
Information). Each SASA value was mean-averaged over four
set of simulations and reported with the appropriate standard
deviation (Figure 4, SASA values are listed under each image).
After 100 ns, the SASAs of the trimer cores were reduced from
the initial ∼1620 Å2 value to 469, 347, and 252 Å2 for cage-T0-
T0, cage-T3-T3, and cage-T6-T6, respectively. From the final
structures (Figure 4), it is clear cage-T3-T3 and cage-T6-T6
assemblies are better able in reducing the initial core SASA (by
79−86%) than cage-T0-T0 (by only 71%) because they have
non-complementary thymine spacers that can wrap around the
cores (Figure 4E,F) to reduce their SASA. This is not the case
for cage T0-T0, which was forced to dehybridize the terminal
base pairs so that the cores can interact with the individual
bases and minimize its exposed hydrophobic surface(Figure
4D). This, however, distorts and destabilizes the DNA duplexes
from their stable B-form conformation as reflected in the
decreased total number of hydrogen bonds (see discussion
below).
At the beginning of the simulations, the total number of H-
bonds within the hybridized DNA duplex arms for all cage
systems was calculated to be 69. However, 13 of these bonds
were broken in cage-T0-T0 after 100 ns, which corresponds to
dehybridization of 6−7 base pairs out of 30 (Figure 4, number
of H-bond are listed under each image). While some of the
initial base pairings broke apart in cage-T3-T3, the number of
H-bonds was still 69 due to formation of additional non-
Watson−Crick hydrogen bonds from non-complementary
thymine spacers. The formation of these non-traditional H-
bonds was more evident in cage-T6-T6, where the total number
of H-bonds increased from 69 to 79. These results suggest that
non-complementary thymine spacers can play important roles
in the stabilization of the cage structures by minimizing the
exposed hydrophobic surfaces of the cores. In the absence of
such spacers, such as in the case of cage-T0-T0, the cage
structure is so severely disturbed to compensate for the exposed
hydrophobic organic cores that it cannot even form; instead, ill-
defined networks are favored, as observed in our experiments.
We note that Sleiman and co-workers recently observed the
exclusive formation of an insoluble ill-defined network when
they combined two “fully complementary” rSMDH3's, where
the ss-DNA arms are directly attached to a tetraphenyl core
without any spacer in between, at 3 μM concentration.20 The
large size of these networks was confirmed by their zero
electrophoretic mobility in PAGE-gel, and they can be
visualized as infinite networks by AFM. These results are
identical to what we observed for our cage-T0-T0 dimer and
reinforced our observations that ill-defined networks are greatly
Figure 4. Molecular-surface representation of cage dimers before (A−
C) and after (D−F) 100 ns MD simulations: cage-T0-T0 (A, D), cage-
T3-T3 (B, E), and cage-T6-T6 (C, F). Final structures are taken from
MD simulations with lowest SASA values. Green, organic core; blue,
non-complementary thymine spacers; orange, 10-bp complementary
ss-DNA arms that can form ds-DNA duplexes. We note that although
the final lowest-SASA structure of cage-T0-T0 (D) has the organic
cores wrapped within the DNA arms, the DNA duplexes are
significantly distorted as evidenced by the large average root-mean-
square deviation (rmsd) of DNAs from B-form conformation (5.8 Å,
see also Table S7 in the Supporting Information).
7453
dx.doi.org/10.1021/ja300322a | J. Am. Chem. Soc. 2012, 134, 7450−7458

Journal of the American Chemical Society
Article
Figure 5. Schematic representation of cage (A) and face-to-face (B) dimers being formed from rSMDH3 building blocks.
preferred over discrete [rSMDH3:rSMDH3] cages in the
absence of non-complementary bases.
rSMDH3 building blocks would “push” the cores together after
the first strands hybridize, the cores may be in close enough
proximity for hydrophobic interaction and possible stacking
(Figure 5B) that leads to exclusive formation of a face-to-face
dimer. We predict that such an assembly would have enhanced
melting temperature compared to simple DNA duplexes as well
as comparable cage dimers because it can take advantage of
both hydrophobic core-stacking and DNA hybridization.
Molecular Dynamics Simulations of Face-to-Face
Dimers. To evaluate the stability of the aforementioned ff
dimer structures, we carried out implicit solvent MD
simulations on model systems representing ff-T0-T0, ff-T3-T0,
and ff-T6-T0 dimer structures using the same methodology as
discussed above for the cage dimers. As before, initial structures
of the ff dimers were constructed with three 10-bp ds-DNA
arms attached to the organic cores, which were placed in the
model as far away as possible from the DNA bases and base
pairs (Figure 6A−C) at the beginning of the simulation. All
final structures after MD simulations (100 ns) exhibited a
planar arrangement of all the hybridized duplex arms (Figure
6D−F), with the cores in each structure either stacked directly
on top of each other (Figure 7) or in close proximity at an
average center-of-mass distance between 5 and 13 Å (Table S6
in the Supporting Information). These organic cores either
intercalated into the base pairs (see Figure 6A for ff-T0-T0) or
were wrapped by non-complementary thymine spacers (see
New Design for Discrete Face-to-Face Dimers That
Take Advantage of Hydrophobic Interactions. Although
hydrophobic interaction has not been previously identified as a
major parameter that governed the assembly of rSMDH-based
assemblies, it has been shown to be the primary interaction that
induces the formation of supramolecular hairpin assemblies
based on hydrophobic stacking of PDI cores.16,18 With this in
mind, we then set out to build a discrete [rSMDH3:rSMDH3]
assembly that take advantage of both the hydrogen-bonding
capability of the DNA-duplex arms and the hydrophobic
interaction between the organic cores. Our experimental
observations and MD simulations clearly demonstrated that
the selectivity for cage vs ill-defined network formation is
controlled not only by the concentration of the rSMDH3
building blocks but also by how well the final assembled
structure can minimize exposure of the hydrophobic surfaces of
the organic cores. In the assembly of our cage dimer, after the
first DNA arms are hybridized together from the two
complementary rSMDH3 building blocks, the two cores are
spaced far apart from each other and thus have the potential to
form linkages that “grow outward” to form a network at higher
concentration (Figure 5A). If the DNA sequences and their
orientations with respect to the core can be redesigned in such
a way that the assembly process from the two complementary
7454
dx.doi.org/10.1021/ja300322a | J. Am. Chem. Soc. 2012, 134, 7450−7458

Journal of the American Chemical Society
Article
As in the case of the caged dimers, the total number of H-
bonds within the hybridized DNA duplex arms in the ff dimers
69 at the start of the simulation. The final ff-T3-T0 and ff-T6-T0
structures did not deviate significantly from this number (64
and 68 H-bonds, respectively), suggesting that there is no
significant rearrangement of the DNA arms to minimize
exposure of the hydrophobic surfaces of the cores. Surprisingly,
while the ff-T0-T0 dimer has lost 14 H-bonds from the
hybridized duplex arms, its SASA value remains much lower
than that of the other two ff dimers, suggesting that this H-
bond lost did not destabilize the structure at all, in stark
contrast to that observed in the cage-T0-T0 dimer (see Figure 4
and accompanying discussion above). This enhanced stability
can be directly attributed to the unique face-to-face
conformation of the two organic cores in ff-T0-T0 dimer,
which benefits not only from the hydrophobic stacking of the
cores (Figure 7) but also from an effective lengthening of the
duplex arms: the stacked core−core units essentially linked two
10-bp duplexes together to create an effectively longer DNA
duplex system (Figure 6D). In comparison to the high SASA
values for the ff-T3-T0 and ff-T6-T0 dimers, the lower SASA
value for the ff-T0-T0 structure suggests that the hydrophobic
stacking of the cores (Figure 7) is the more important
contributor to this stabilization. Indeed, while the core−core
stacking still occurs for ff-T3-T0 and ff-T6-T0, the non-
complementary thymine spacers can better wrap around the
stacked core−core unit to reduce the SASA as well as the
hydrophobic stacking interactions between the cores (Figure
6E,F). Indeed, MD simulations indicated that the average
center-of-mass core-to-core distance in four sets of each ff
dimer (5, 11, and 13 Å for ff-T0-T0, ff-T3-T0, and ff-T6-T0,
respectively, see also Table S7 in the Supporting Information)
increases as the spacer length increases
Figure 6. Molecular-surface representation of face-to-face dimers
before (A−C) and after (D−F) 100 ns MD simulations: ff-T0-T0 (A,
D), ff-T3-T3 (B, E), and ff-T6-T6 (C, F). Final structures are taken
from MD simulations with lowest SASA values. Side views are shown
in B and C to emphasize the 3D structures. Top views of the structures
are shown in A, D, E, and F. Green, organic core; blue, non-
complementary thymine spacers; orange, 10-bp complementary ss-
DNA arms that can form ds-DNA duplexes.
Self-Assembly of Face-to-Face Dimers. We were
pleasantly surprised to observe exclusive formation of stable ff
dimer structures (Table 3) from combining rSMDH3-XT0-3′
Table 3. Face-to-Face Dimers Based on rSMDH3's
a
entry
components
short name
1
2
3
rSMDH3-X′T0-5′:rSMDH3-XT0-3′
rSMDH3-X′T3-5′:rSMDH3-XT0-3′
rSMDH3-X′T6-5′:rSMDH3-XT0-3′
ff-T0-T0
ff-T3-T0
ff-T6-T0
a
Total DNA concentration of 1 and 2 μM.
Figure 7. Top and side views of the hydrophobic cores being stacked
on top of each other in the ff-T0-T0 final structure. Light and dark
green-colored molecular surfaces distinguish the individual cores.
and its complementary rSMDH3-X′Tn-5′ (n = 0, 3, and 6) at 2
μM total ss-DNA concentration. The non-denaturing PAGE-
gel for these ff dimers shows one single clean band for each of
these dimers (Figure 8). As the MD simulation suggested, and
as shown by the optical melting profiles of these ff dimers
(Figure 9), the hydrophobic core−core stacking in these
assemblies resulted in dramatic increases in their melting
temperatures (Tm’s) compared to the control-rSMDH3-T3 and
cage-T3-T3 dimer: ff-T0-T0 exhibited a dramatic 18.4 °C
increase in Tm over the control-rSMDH3-T3 assembly and a
14 °C increase in Tm over that for cage-T3-T3 dimer. The extra
stability in ff-T0-T0 can be explained by the interactions of the
stacked cores with the closest base pairs of the DNA duplex
arms, similar to that observed in short DNA duplexes
possessing hydrophobic end-capping agents.52
Figures 6E for ff-T3-T0 and ff-T6-T0 and 6F) to minimize
exposure of the hydrophobic surfaces of the cores to the
aqueous solvent.
SASA analysis for all three ff dimers showed that they are all
in the range of 199−285 Å2 (Figure 6, SASA values are listed
under each image), which is lower than those found for the
experimentally observed cage-T6-T6 and cage-T3-T3 dimers
(252−347 Å2, Figure 4, SASA values are listed under each
image). These lower SASA values can be attributed to the
hydrophobic interactions between the cores observed in the
simulations, which help to minimize the exposure of the cores
to the aqueous media without significantly disturbing the DNA
duplexes, and suggests that the ff dimers would be stable and
can be observed experimentally.
As we introduced T3 and T6 spacers into the rSMDH3-X′Tn-
5′ partner and use these to form ff-T3-T0 and ff-T6-T0 dimers,
the melting temperatures of these dimers successively
7455
dx.doi.org/10.1021/ja300322a | J. Am. Chem. Soc. 2012, 134, 7450−7458

Journal of the American Chemical Society
Article
Table 4. Melting Data for cage-Tn-Tn, control-rSMDH3-Tn,
and ff-Tn-T0 (n = 0, 3, and 6) Assemblies with 1 μM Total
DNA Concentration in PBS Buffer (10 mM, pH 7.0, 75 mM
NaCl)
entry
short name
X:X′
Tm (°C)
fwhm (°C)
1
36.6 0.2
39.2 0.3
36.5 0.2
37.4 0.1
41.2 0.3
39.8 0.3
39.6 0.2
40.6 0.1
55.6 0.1
50.3 0.3
46.0 0.2
12.0 0.1
11.4 0.1
9.9 0.1
9.3 0.1
9.2 0.1
7.0 0.1
7.4 0.1
7.3 0.1
7.1 0.1
7.2 0.1
7.4 0.1
2
control-rSMDH3-T0
control-rSMDH3-T3
control-rSMDH3-T6
3
4
a
5
“cage-T0-T0”
6
cage-T3-T3
cage-T6-T6
cage-T3-T3-reverse
ff-T0-T0
7
8
9
10
11
ff-T3-T0
ff-T6-T0
a
This structure did not form according to PAGE-gel study (see Figure
3).
quenching behavior has been observed in PDI-based hairpins
as they formed supramolecular polymers based on stacking of
PDI cores in high salt environment.18 Indeed, the fluorescence
of the three ff dimers (Figure 10) varies inversely with their
Figure 8. Non-denaturing PAGE-gel image (6%) of DNA hybrids (2
μM). From left to right: lane 1 = ff-T0-T0, lane 2 = ff-T3-T0, lane 3 = ff-
T6-T0.
Figure 10. Fluorescent spectra of ff-T0-Tn (n = 0, 3, and 6) and cage-
T3-T3 (1 μM) dimers in PBS buffer (10 mM, pH 7.0, 75 mM NaCl);
the excitation wavelength was 305 nm. The fluorescent spectrum of
the cage-T3-T3 assembly is provided as a reference point rather than
that for the [rSMDH3-XT0-3′:rSMDH3-X′T0-5′] assembly (Figure S34
in the Supporting Information) because quenching of the core by
conformationally fixed DNA base pairs does occur to a certain
extent.79
Figure 9. Melting profiles for ff-Tn-T0 (n = 0, 3, and 6) and cage-T3-T3
(1 μM) in PBS buffer (10 mM, pH 7.0, 75 mM NaCl).
decreased from that of the ff-T0-T0 dimer, by 4 and 8 °C,
respectively (Figure 9 and Table 4). Nevertheless, these are still
significantly higher than those observed for the corresponding
control-rSMDH3-Tn assemblies and cage-Tn-Tn dimers (n = 0,
3, and 6; see Table 4). This trend confirms our observation in
the MD simulations (see discussion above) that the T3 and T6
spacers can interact with the stacked core−core unit and
decrease the hydrophobic stacking interactions that are
primarily responsible for stabilizing the ff-T0-T0 dimer (see
C−C distance analysis in Table S7 in the Supporting
Information).
stability, which in turn is a function of the distance between the
cores. Most interestingly, the fluorescence-based melting profile
of ff-T0-T0 closely mirrors its optical melting profile (Figure
S37 in the Supporting Information), indicating that in the
melting of the ff dimers, core−core unstacking occurs in a
concerted manner as the dehybridization of the duplex
arms.53,78
Further evidence for the hydrophobic stacking of the cores in
ff dimers is provided by the observation that the fluorescence of
the core in rSMDH3-XT0-3′ and its complementary rSMDH3-
X′T0-5′ partner is greatly quenched upon hybridization into the
ff-T0-T0 (Figure S34 in the Supporting Information) due to the
increased interactions between the cores. This type of
Enhanced Melting Properties of Cage and Face-to-
Face Dimers. We have previously reported that cage dimers
based on rSMDH223 and rSMDH321 building blocks can display
cooperative melting behaviors (i.e., increased Tm and sharpened
melting profiles) in comparison to their free DNA analogues
due to a combination of reduced configurational entropy and
7456
dx.doi.org/10.1021/ja300322a | J. Am. Chem. Soc. 2012, 134, 7450−7458

Journal of the American Chemical Society
Article
ion-cloud sharing that occurs as a consequence of the proximity
of their DNA duplex linkages. To these two parameters of the
neighboring-duplex model28 can now be added hydrophobic
interactions, especially if the organic cores in rSMDHn are
designed to be in the right orientation so that these
interactions, and hydrophobic stacking, can play an important
role, as in the case of ff-T0-T0. Indeed, the three ff-Tn-T0 (n = 0,
3, and 6) dimers reported herein all showed cooperative
melting behaviors that are similar to those of the cage-Tn-Tn (n
= 3 and 6) dimers. The full-widths at half-max (fwhm’s) of the
derivative of the melting profiles for ff-Tn-T0 (n = 0, 3, and 6)
and cage-Tn-Tn (n = 3 and 6) dimers are all in the range of 7.0−
7.4 °C, much sharper than those for the control-rSMDH3-Tn
systems (fwhm ∼9.3−11.4 °C, see Table 4). This sharp melting
behavior can be attributed to configurational entropy23,57 and
ion-cloud sharing between the DNA-duplex arms, all of which
are in the 10−50 Å range required for ion-cloud sharing.80
The Tm of the ff-T0-T0 dimer is dramatically higher
compared to those for the control-rSMDH3-T0 and cage-T3-
T3 dimer (ΔTm = 16.4 and 15.8 °C, respectively, Table 4).
However, as non-complementary T3 and T6 spacers were
introduced into the ff dimers, the melting temperatures of ff-T3-
T0 and ff-T6-T0 were decreased by 5.3 and 9.6 °C compared to
that of the ff-T0-T0 dimer. Yet, ff-T6-T0 still melts at a higher
temperature than the cage-T6-T6 dimer (ΔTm = 6.4 °C). These
observations indicate that a large contribution to the extra
thermal stability in the ff dimers stemmed from the core−core
hydrophobic interactions that can be tuned by the length of the
non-complementary thymine spacers. When there is no spacer,
as in ff-T0-T0, this thermal stability can be further enhanced by
the additional interaction of the stacked cores with the closest
base-pairs from the DNA duplex arms, similar to that observed
in duplexes with hydrophobic end-capping agents.52 This was
indeed observed in our MD simulations (Figure 6D).
ASSOCIATED CONTENT
* Supporting Information
■
S
Synthesis of phosphoramidite core and rSMDHs; HPLC traces
and MALDI-ToF spectra of rSMDHs; optical melting profiles,
first derivatives of the melting profile, fwhm, and error analysis
of all assemblies; fluorescent spectra of ff dimers and their
rSMDH precursors; fluorescence-based melting profiles of ff-
T0-T0 dimer; and molecular dynamics simulations and analysis
details. Complete refs 60, 61, 62, and 65. This material is
available free of charge via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
schatz@chem.northwestern.edu; stn@northwestern.edu
■
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We thank Prof. Frederick D. Lewis for insightful discussions
regarding the fluorescence self-quenching experiments. Dr. Ilyas
Yildirim thanks the Faculty of Science, Akdeniz University
(Antalya, Turkey) for providing him with a working environ-
ment during the preparation of this manuscript. Financial
support for this work was provided by the Air Force Office of
Scientific Research under agreement FA-9550-11-1-0275 and
the NSF (Grant EEC-0647560 through the NSEC program).
REFERENCES
■
(1) Bloomfield, V. A.; Crothers, D. M.; Tinoco, I. Nucleic Acids:
Structures, Properties, and Functions; University Science Books:
Sausalito, CA, 2000.
(2) Seeman, N. C. Nature 2003, 421, 427−431.
(3) Lin, C.; Liu, Y.; Rinker, S.; Yan, H. ChemPhysChem 2006, 7,
1641−1647.
(4) Lin, C.; Liu, Y.; Yan, H. Biochemistry 2009, 48, 1663−1674.
(5) Rothemund, P. W. K. Nature 2006, 440, 297−302.
(6) Zhang, C.; He, Y.; Su, M.; Ko, S. H.; Ye, T.; Leng, Y.; Sun, X.;
Ribbe, A. E.; Jiang, W.; Mao, C. Faraday Discuss. 2009, 143, 221−233.
(7) Lo, P. K.; Metera, K. L.; Sleiman, H. F. Curr. Opin. Chem. Biol.
2010, 14, 597−607.
(8) He, Y.; Ye, T.; Su, M.; Zhang, C.; Ribbe, A. E.; Jiang, W.; Mao, C.
Nature 2008, 452, 198−201.
(9) Andersen, E. S.; Dong, M.; Nielsen, M. M.; Jahn, K.; Subramani,
R.; Mamdouh, W.; Golas, M. M.; Sander, B.; Stark, H.; Oliveira, C. L.
P.; Pedersen, J. S.; Birkedal, V.; Besenbacher, F.; Gothelf, K. V.; Kjems,
J. Nature 2009, 459, 73−76.
CONCLUSION
■
In summary, we have elucidated the important role that
hydrophobic interactions play in the aqueous assembly of
nanoscale organic-DNA hybrid cages and ff dimers from
rSMDHn building blocks. These nanostructures are assembled
in aqueous environments in such a way that minimizes the
exposure of the hydrophobic surface of the organic cores,
whether through hydrophobic interactions with the core in
another rSMDHn, with the bases in the attached DNA strands,
and/or with the base pairs in the final assembled structures.
When exposure of the hydrophobic surfaces of the cores could
not be properly minimized in a dimer conformation, an ill-
defined network can result, as in the case of cage-T0-T0 dimer
made from rSMDH3 building blocks. On the other hand, if a
dimer can be formed with good minimization of the
hydrophobic surfaces of the cores, highly stable structures
with enhanced melting points and cooperative melting profiles
can be obtained. When hydrophobic stacking between the cores
is the dominant interaction, such as in the case of ff dimers,
dramatic increases in the melting temperature can be obtained.
In the world of nanoscale DNA-linked structures, these results
place hydrophobic interactions at the same level of importance
as the number of ss-DNA surrounding the organic core; the
orientation, geometry, and concentration of the strands; and
the type and concentration of the salts in the assembly media.
We are confident that such interactions can be utilized
effectively as a design parameter in future applications of
DNA-linked materials.
(10) Le, J. D.; Pinto, Y.; Seeman, N. C.; Musier-Forsyth, K.; Taton,
T. A.; Kiehl, R. A. Nano Lett. 2004, 4, 2343−2347.
(11) Zheng, J.; Constantinou, P. E.; Micheel, C.; Alivisatos, A. P.;
Kiehl, R. A.; Seeman, N. C. Nano Lett. 2006, 6, 1502−1504.
(12) Li, H.; Park, S. H.; Reif, J. H.; LaBean, T. H.; Yan, H. J. Am.
Chem. Soc. 2003, 126, 418−419.
(13) Yan, H.; Park, S. H.; Finkelstein, G.; Reif, J. H.; LaBean, T. H.
Science 2003, 301, 1882−1884.
(14) Liu, Y.; Lin, C.; Li, H.; Yan, H. Angew. Chem., Int. Ed. 2005, 44,
4333−4338.
(15) Shi, J.; Bergstrom, D. E. Angew. Chem., Int. Ed. 1997, 36, 111−
113.
(16) Wang, W.; Wan, W.; Zhou, H.-H.; Niu, S.; Li, A. D. Q. J. Am.
Chem. Soc. 2003, 125, 5248−5249.
(17) Abdalla, M. A.; Bayer, J.; Radler, J. O.; Mullen, K. Angew. Chem.,
̈
̈
Int. Ed. 2004, 43, 3967−3970.
(18) Neelakandan, P. P.; Pan, Z.; Hariharan, M.; Zheng, Y.;
Weissman, H.; Rybtchinski, B.; Lewis, F. D. J. Am. Chem. Soc. 2010,
132, 15808−15813.
7457
dx.doi.org/10.1021/ja300322a | J. Am. Chem. Soc. 2012, 134, 7450−7458

Journal of the American Chemical Society
Article
(19) Kuroda, T.; Sakurai, Y.; Suzuki, Y.; Nakamura, A. O.; Kuwahara,
M.; Ozaki, H.; Sawai, H. Chem.−Asian J. 2006, 1, 575−580.
(20) Aldaye, F. A.; Sleiman, H. F. J. Am. Chem. Soc. 2007, 129,
10070−10071.
(52) Ng, P.-S.; Laing, B. M.; Balasundarum, G.; Pingle, M.; Friedman,
A.; Bergstrom, D. E. Bioconjugate Chem. 2010, 21, 1545−1553.
(53) Hariharan, M.; Zheng, Y.; Long, H.; Zeidan, T. A.; Schatz, G. C.;
Vura-Weis, J.; Wasielewski, M. R.; Zuo, X.; Tiede, D. M.; Lewis, F. D.
J. Am. Chem. Soc. 2009, 131, 5920−5929.
(21) Stepp, B. R.; Gibbs-Davis, J. M.; Koh, D. L. F.; Nguyen, S. T. J.
Am. Chem. Soc. 2008, 130, 9628−9629.
(54) Andrews, A. T. Electrophoresis: Theory, Techniques, and
Biochemical and Clinical Applications, 2nd ed.; Clarenden Press:
Oxford, 1986.
(22) Meng, M.; Ahlborn, C.; Bauer, M.; Plietzsch, O.; Soomro, S. A.;
Singh, A.; Muller, T.; Wenzel, W.; Brase, S.; Richert, C. ChemBioChem
2009, 10, 1335−1339.
(23) Eryazici, I.; Prytkova, T. R.; Schatz, G. C.; Nguyen, S. T. J. Am.
Chem. Soc. 2010, 132, 17068−17070.
(24) Singh, A.; Tolev, M.; Meng, M.; Klenin, K.; Plietzsch, O.;
(55) Knotts, T. A.; Rathore, N.; Schwartz, D. C.; de Pablo, J. J. J.
Chem. Phys. 2007, 126, 084901.
(56) Drukker, K.; Schatz, G. C. J. Phys. Chem. B 2000, 104, 6108−
6111.
(57) Prytkova, T. R.; Eryazici, I.; Stepp, B.; Nguyen, S. B.; Schatz, G.
C. J. Phys. Chem. B 2010, 114, 2627−34.
Schilling, C. I.; Muller, T.; Nieger, M.; Brase, S.; Wenzel, W.; Richert,
̈
C. Angew. Chem., Int. Ed. 2011, 50, 3227−3231.
̆ ̆
(58) Sayar, M.; Avsa̧ roglu, B.; Kabakcı̧ oglu, A. Phys. Rev. E 2010, 81,
(25) Shchepinov, M. S.; Udalova, I. A.; Bridgman, A. J.; Southern, E.
M. Nucleic Acids Res. 1997, 25, 4447−4454.
041916.
(59) Aldaye, F. A.; Sleiman, H. F. Pure Appl. Chem. 2009, 81, 2157−
2181.
(60) Case, D. A. et al. AMBER 9; University of California, San
(26) Scheffler, M.; Dorenbeck, A.; Jordan, S.; Wustefeld, M.; von
̈
Kiedrowski, G. Angew. Chem., Int. Ed. 1999, 38, 3311−3315.
(27) Shchepinov, M. S.; Mir, K. U.; Elder, J. K.; Frank-Kamenetskii,
M. D.; Southern, E. M. Nucleic Acids Res. 1999, 27, 3035−3041.
(28) Gibbs-Davis, J. M.; Schatz, G. C.; Nguyen, S. T. J. Am. Chem.
Soc. 2007, 129, 15535−15540.
Francisco, 2006.
(61) Case, D. A. et al. AMBER 10; University of California, San
Francisco, 2006.
(62) Case, D. A. et al. AMBER 11; University of California, San
Francisco, 2006.
(63) MacKerell, A. D., Jr.; Banavali, N.; Foloppe, N. Biopolymers
2000, 56, 257−265.
(64) Hart, K.; Foloppe, N.; Baker, C. M.; Denning, E. J.; Nilsson, L.;
MacKerell, A. D., Jr. J. Chem. Theory Comput. 2011, 8, 348−362.
(65) Shaw, D. E.; et al. SIGARCH Comput. Archit. News 2007, 35, 1−
12.
(29) Alemdaroglu, F. E.; Ding, K.; Berger, R.; Herrmann, A. Angew.
Chem., Int. Ed. 2006, 45, 4206−4210.
(30) Ding, K.; Alemdaroglu, F. E.; Borsch, M.; Berger, R.; Herrmann,
̈
A. Angew. Chem., Int. Ed. 2007, 46, 1172−1175.
(31) Waybright, S. M.; Singleton, C. P.; Tour, J. M.; Murphy, C. J.;
Bunz, U. H. F. Organometallics 2000, 19, 368−370.
(32) Waybright, S. M.; Singleton, C. P.; Wachter, K.; Murphy, C. J.;
Bunz, U. H. F. J. Am. Chem. Soc. 2001, 123, 1828−1833.
(33) Vargas-Baca, I.; Mitra, D.; Zulyniak, H. J.; Banerjee, J.; Sleiman,
H. F. Angew. Chem., Int. Ed. 2001, 40, 4629−4632.
(34) Stewart, K. M.; McLaughlin, L. W. Chem. Commun. 2003,
2934−2935.
(66) Klepeis, J. L.; Lindorff-Larsen, K.; Dror, R. O.; Shaw, D. E. Curr.
Opin. Struct. Biol. 2009, 19, 120−127.
(67) Wang, J.; Wolf, R. M.; Caldwell, J. W.; Kollman, P. A.; Case, D.
A. J. Comput. Chem. 2004, 25, 1157−1174.
(68) Cornell, W. D.; Cieplak, P.; Bayly, C. I.; Kollmann, P. A. J. Am.
Chem. Soc. 1993, 115, 9620−9631.
(35) Mitra, D.; Di Cesare, N.; Sleiman, H. F. Angew. Chem., Int. Ed.
2004, 43, 5804−5808.
(69) Cornell, W. D.; Cieplak, P.; Bayly, C. I.; Gould, I. R.; Merz, K.
M.; Ferguson, D. M.; Spellmeyer, D. C.; Fox, T.; Caldwell, J. W.;
Kollman, P. A. J. Am. Chem. Soc. 1995, 117, 5179−5197.
(70) Bayly, C. I.; Cieplak, P.; Cornell, W.; Kollman, P. A. J. Phys.
Chem. 1993, 97, 10269−10280.
(36) Stewart, K. M.; McLaughlin, L. W. J. Am. Chem. Soc. 2004, 126,
2050−2057.
(37) Stewart, K. M.; Rojo, J.; McLaughlin, L. W. Angew. Chem., Int.
Ed. 2004, 43, 5808−5811.
(38) Mirkin, C. A.; Letsinger, R. L.; Mucic, R. C.; Storhoff, J. J.
Nature 1996, 382, 607−609.
(71) Cieplak, P.; Cornell, W. D.; Bayly, C.; Kollman, P. A. J. Comput.
Chem. 1995, 16, 1357−1377.
(39) Alivisatos, A. P.; Johnsson, K. P.; Peng, X.; Wilson, T. E.;
Loweth, C. J.; Bruchez, M. P.; Schultz, P. G. Nature 1996, 382, 609−
611.
(72) Wang, J.; Cieplak, P.; Kollman, P. A. J. Comput. Chem. 2000, 21,
1049−1074.
(73) Yildirim, I.; Stern, H. A.; Kennedy, S. D.; Tubbs, J. D.; Turner,
D. H. J. Chem. Theory Comput. 2010, 6, 1520−1531.
(40) Elghanian, R.; Storhoff, J. J.; Mucic, R. C.; Letsinger, R. L.;
Mirkin, C. A. Science 1997, 277, 1078−1081.
(41) Taton, T. A.; Mirkin, C. A.; Letsinger, R. L. Science 2000, 289,
1757−1760.
́ ́
(74) Perez, A.; Marchan, I.; Svozil, D.; Sponer, J.; Cheatham, T. E.,
III; Laughton, C. A.; Orozco, M. Biophys. J. 2007, 92, 3817−3829.
(75) Hawkins, G. D.; Cramer, C. J.; Truhlar, D. G. J. Phys. Chem.
1996, 100, 19824−19839.
(42) Gibbs, J. M.; Park, S.-J.; Anderson, D. R.; Watson, K. J.; Mirkin,
C. A.; Nguyen, S. T. J. Am. Chem. Soc. 2005, 127, 1170−1178.
(43) Park, S. Y.; Lytton-Jean, A. K. R.; Lee, B.; Weigand, S.; Schatz,
G. C.; Mirkin, C. A. Nature 2008, 451, 553−556.
(44) Aldaye, F. A.; Sleiman, H. F. Angew. Chem., Int. Ed. 2006, 45,
2204−2209.
(76) Hawkins, G. D.; Cramer, C. J.; Truhlar, D. G. Chem. Phys. Lett.
1995, 246, 122−129.
(77) Collaborative Computational Project, Number 4. Acta
Crystallogr., Sect. D 1994, 50, 760−763.
(78) Hariharan, M.; Siegmund, K.; Zheng, Y.; Long, H.; Schatz, G.
C.; Lewis, F. D. J. Phys. Chem. C 2010, 114, 20466−20471.
(79) Lewis, F. D.; Letsinger, R. L.; Wasielewski, M. R. Acc. Chem. Res.
2001, 34, 159−170.
(45) Aldaye, F. A.; Sleiman, H. F. J. Am. Chem. Soc. 2007, 129, 4130−
4131.
(46) Lo, P. K.; Karam, P.; Aldaye, F. A.; McLaughlin, C. K.; Hamblin,
G. D.; Cosa, G.; Sleiman, H. F. Nat. Chem. 2010, 2, 319−328.
(47) Macfarlane, R. J.; Lee, B.; Jones, M. R.; Harris, N.; Schatz, G. C.;
Mirkin, C. A. Science 2011, 334, 204−208.
(48) Yang, H.; Sleiman, H. F. Angew. Chem., Int. Ed. 2008, 47, 2443−
2446.
(80) Long, H.; Kudlay, A.; Schatz, G. C. J. Phys. Chem. B 2006, 110,
2918−2926.
(49) Yang, H.; McLaughlin, C. K.; Aldaye, F. A.; Hamblin, G. D.;
Rys, A. Z.; Rouiller, I.; Sleiman, H. F. Nat. Chem. 2009, 1, 390−396.
(50) Dogan, Z.; Paulini, R.; Stutz, J. A. R.; Narayanan, S.; Richert, C.
̈
J. Am. Chem. Soc. 2004, 126, 4762−4763.
(51) Patra, A.; Richert, C. J. Am. Chem. Soc. 2009, 131, 12671−
12681.
7458
dx.doi.org/10.1021/ja300322a | J. Am. Chem. Soc. 2012, 134, 7450−7458