Content uploaded by Hanadi F Sleiman
Author content
All content in this area was uploaded by Hanadi F Sleiman on Sep 09, 2014
Content may be subject to copyright.
This journal is cThe Royal Society of Chemistry 2011 Chem. Soc. Rev., 2011, 40, 5647–5656 5647
Cite this:
Chem. Soc. Rev
., 2011, 40, 5647–5656
Supramolecular DNA assemblyw
Christopher K. McLaughlin, Graham D. Hamblin and Hanadi F. Sleiman*
Received 16th September 2011
DOI: 10.1039/c1cs15253j
The powerful self-assembly features of DNA make it a unique template to finely organize and
control matter on the nanometre scale. While DNA alone offers a high degree of fidelity in its self-
assembly, a new area of research termed ‘supramolecular DNA assembly’ has recently emerged.
This field combines DNA building blocks with synthetic organic, inorganic and polymeric
structures. It thus brings together the toolbox of supramolecular chemistry with the predictable and
programmable nature of DNA. The result of this molecular partnership is a variety of hybrid
architectures, that expand DNA assembly beyond the boundaries of Watson–Crick base pairing
into new structural and functional properties. In this tutorial review we outline this emerging field of
study, and describe recent research aiming to synergistically combine the properties inherent to
DNA with those of a number of supramolecular scaffolds. This ultimately creates structures with
numerous potential applications in materials science, catalysis and medicine.
1. Introduction
The search for methods to pattern molecules and materials
with nanometre scale precision has been one of the most
intellectually charged objectives in the last few decades. From
continually pushing the limits of top-down approaches, to
finding ever new ways to organize components from the
bottom-up, chemists, physicists, biologists, engineers and
computer scientists have joined forces to tame assembly down
to the smallest size scale, and to learn new science in the process.
Natural systems have provided both muse and muscle for this
field, having long evolved effective solutions to organize complex
architectures with sophisticated function. Of the naturally
evolved systems, DNA is a fascinating example of self-assembly;
widely recognized as the carrier of our genetic code, DNA has
Department of Chemistry, McGill University, 801 Sherbrooke St.
West, Montreal, QC H3A 2K6, Canada.
E-mail: hanadi.sleiman@mcgill.ca; Fax: +1 514 398 3797;
Tel: +1 514 398 2633
wPart of a themed issue on the advances in DNA-based nanotechnology.
Graham D. Hamblin, Hanadi F. Sleiman and
Christopher K. McLaughlin
Christopher K. McLaughlin (right) grew up in the picturesque
village of Tamworth, Ontario. He studied chemistry at the University
of Guelph (Guelph, Canada) and received his BSc in Chemistry in
2004. He continued on at Guelph and obtained an MSc studying the
photophysical properties of modified purine nucleosides under the
supervision of Prof. Richard Manderville in 2006. He then started his
PhD studies at McGill University in 2006 under the supervision of
Prof. H. Sleiman and works on developing methods to access
functionally addressable 3D DNA nanostructures.
Graham Hamblin (left) grew up in Calgary, Alberta, and received
his BSc in Chemistry from the University of Calgary in 2008. He is a
PhD candidate at McGill University working under the supervision of
Prof. H. Sleiman, with research primarily focused on the design and
applications of DNA nanotubes.
Hanadi Sleiman (middle) received her PhD from Stanford
University under the guidance of Prof. L. McElwee-White. Following
a CNRS postdoctoral stay in supramolecular chemistry with
Prof. Jean-Marie Lehn at the Universite
´Louis Pasteur in France, she joined the faculty of McGill University in 1999, where she
is currently professor of chemistry and Dawson Scholar (McGill’s Canada Research Chair). The Sleiman research group focuses on
developing the supramolecular chemistry of DNA, towards applications in biology and in nanoscience. Sleiman was named Cottrell
Scholar of the Research Corporation in 2002. She received the Principal’s Prize (2002) and the Leo Yaffe Award (2004) for
excellence in teaching, the NSERC Discovery Accelerator Award in 2008, and the 2009 Strem Award of the Canadian Society for
Chemistry.
Chem Soc Rev Dynamic Article Links
www.rsc.org/csr TUTORIAL REVIEW
5648 Chem. Soc. Rev., 2011, 40, 5647–5656 This journal is cThe Royal Society of Chemistry 2011
obvious ties to biology, evolution and medicine. DNA is possibly
the most predictable and programmable self-assembling material
known, undergoing rapid, high fidelity association according
to simple base-pairing rules. Through a cooperative balance of
hydrogen-bonding, p-stacking and other non-covalent forces,
a single strand of DNA can find its complement in solution
with extraordinary selectivity. Though many self-assembling
molecules are known, such predictable control over the resulting
product is rare. DNA gives researchers a precise assembly
code to define which strands will pair, and results in double-
helices that are structurally uniform and well-characterized.
For these reasons, DNA assembly has been exploited in the
field of structural DNA nanotechnology to generate higher-
order functional constructs.
1,2
In parallel, an important and now mature research area that
has significantly contributed to bottom-up construction is
supramolecular chemistry. Over the last forty years, this field
has developed an in-depth understanding of non-covalent
interactions. It has evolved a toolbox of synthetic organic
and inorganic molecules that use these interactions to sponta-
neously assemble into remarkably diverse architectures.
3–5
Because the component molecules can possess intrinsic func-
tionality, such as redox, photophysical, magnetic and catalytic
properties, they allow the rational design of complex function
into the final assemblies. However, while supramolecular
chemistry can successfully organize symmetrical structures
and periodic assemblies, its ability to achieve the programmable
positioning of components into any deliberately designed struc-
ture, whether symmetric or asymmetric, is less well-developed.
Can we bring together the programmability of DNA, with
the structural and functional diversity achieved by supramolecular
chemistry? This is being explored in a new emerging area that
was recently termed ‘supramolecular DNA assembly’.
6
By
blending together DNA building blocks with synthetic organic
and inorganic molecules, the field of DNA nanotechnology
can gain completely new structural motifs and can impart
functionality to the typically passive DNA structures (Fig. 1).
By adding DNA to its toolbox, the field of supramolecular
chemistry acquires the ability to position components into any
programmable pattern, in order to engineer sophisticated
function. Thus, this new area is poised to enrich both fields
with completely new structures and new applications in both
biology and materials science.
In this tutorial review, we outline seminal and recent
research in this exciting field, specifically focusing on the
underlying advantages of synthetic insertions in two- and
three-dimensional nanostructures, with the intent to provide
a palatable introduction for the non-expert. For more extensive
coverage of specific areas of interest, we direct the reader to
topical reviews at the beginning of each sub-section.
2. DNA with synthetic insertions
2.1 Using synthetic vertices to bring higher order to DNA
structures
The simplest chemical insertions into DNA strands are able to
alter its hybridization and control its self-assembly outcome. By
incorporating molecules with specific geometries at the insertion
points, basic linear duplexes can be oriented relative to each
other in a unique manner, combining the self-assembly proper-
ties of DNA with the structural versatility of synthetic groups.
Fig. 1 The union of supramolecular chemistry with structural DNA nanotechnology gives rise to a new area with materials and biological
applications. Reproduced from ref. 28, 40, and 52 with permission from American Chemical Society. Reproduced from ref. 9, 24, and 45 with
permission from Wiley-VCH. Reproduced from ref. 3, 4, and 5 with permission from the Royal Society of Chemistry.
This journal is cThe Royal Society of Chemistry 2011 Chem. Soc. Rev., 2011, 40, 5647–5656 5649
The result is the introduction of new construction motifs for
nanotechnology.
An early example of this is illustrated by the work of
Bergstrom and coworkers,
7
which employs two-arm vertices.
The vertex centers around a single tetrahedral carbon and
is connected to two single-stranded (ss) DNA linkers via rigid
p-(2-hydroxyethyl)phenylethynylphenyl spacer units (Fig. 2a).
In this case, self-complementary ssDNA linkers were included,
so that their hybridization would yield discrete macrocycle
formation. Because any number of monomers can associate,
a ladder of oligomeric products results, with the primary
product a function of monomer concentration. Increasing
flexibility by introducing non-pairing thymine residues
between the DNA-vertex bond disfavoured oligomerization.
Though not emphasized in Bergstrom’s article, this is evidence
that DNA assembly was indeed manipulated by the synthetic
insertion. The group of von Kiedrowski constructed a tris-
branched oligonucleotide containing a flexible vertex and three
identical DNA strands.
8
They showed that it can assemble
under kinetic conditions into a dimeric structure, in which the
vertices are connected by three DNA double helices, along
with other higher-order structures.
With the aim of obtaining a single, well-defined product, our
group developed DNA-conjugated organic vertices, in which
the arms are no longer self-complementary or identical to one
another (Fig. 2b).
9
The m-terphenyl-based chemical insertion
is a significantly more rigid vertex, and linker sequences were
designed to ensure the hexamer was the only product available
that satisfied Watson–Crick base-pairing. The result is a very
narrow product distribution, comprised of predominantly the
DNA hexagon. This system was used as a scaffold for organizing
gold nanoparticles (AuNPs) into a discrete 2D structure (Fig. 2b,
right). The synthetically modified DNA strands were conjugated
to nanoparticles prior to assembly, allowing them to template
the formation of a cyclic hexamer of AuNPs.
Subsequently, we constructed cyclic DNA polygons, composed
of single-stranded DNA arms and rigid m-terphenyl corner
units (Fig. 3).
10
For example, a DNA ss triangle was synthesized
by on-column incorporation of three m-terphenyl insertions
into a DNA strand, followed by templated joining together of
its two extremities by a phosphodiester linkage (Fig. 3a).
Using this simple strategy, ss DNA squares, pentagons and
hexagons were also selectively constructed. These could then
be used as templates for the positioning of materials by
hybridization to their single-stranded arms. For example, we
used these templates to position gold nanoparticles into a
diverse array of polygonal assemblies.
Because of their single-stranded and fully cyclic nature,
these could serve as dynamic templates that allow writing,
erasing, and structurally switching nanoparticle assemblies,
using externally added DNA strands. We use this literature
example to illustrate the phenomenon of strand displacement,
that has emerged as a powerful tool in DNA nanotechnology.
11
If a DNA strand a-b is hybridized to one side of the DNA
triangle, then a duplex will form with a single-stranded over-
hang or ‘toehold’ (Fig. 3i). When a fully complementary DNA
strand a0-b0is added, strand a-b is displaced to yield a fully
complementary, longer duplex. This process occurs rapidly
and in quantitative yields. If strand a-b is conjugated to a gold
nanoparticle and hybridized to one of the triangle arms, the
added fully complementary sequence can remove this strand
thus ‘erasing’ the nanoparticle from the assembly (Fig. 3ii).
A new smaller nanoparticle could be installed by hybridization,
which can change the optoelectronic properties of these assemblies
in a deterministic manner (Fig. 3iii). We also used these templates
Fig. 2 DNA modification with organic vertices. (a) An organic vertex is
incorporated into DNA to prepare bis-DNA building blocks that undergo
macrocycle formation. Reproduced from ref. 7 with permission from
Wiley-VCH. (b) Asymmetric DNA building blocks with organic vertices
are used to prepare discrete hexagons that can assemble gold nano-
particles. Reproduced from ref. 9 with permission from Wiley-VCH.
Fig. 3 Single-stranded DNA polygons. (a) Synthetic vertices are
incorporated into the growing DNA strand on a solid-support,
followed by cleavage and a templated ligation to yield the ssDNA
structure. (b) DNA- AuNP conjugates can be organized onto the
single-stranded scaffold in a discrete manner and subjected to ‘write/
erase’ experiments. Adapted with permission from ref. 10. Copyright
2007 American Chemical Society.
5650 Chem. Soc. Rev., 2011, 40, 5647–5656 This journal is cThe Royal Society of Chemistry 2011
to structurally switch a AuNP tetramer between trapezoid,
square and rectangle shapes. Monoconjugation of DNA to
AuNPs and subsequent self-assembly into well-defined 2D and
3D nanostructures demonstrates the high degree of positional
control that DNA allows.
12
These nanoparticle assemblies are
expected to show applications as dynamic, tailorable nano-
electronic and plasmonic substrates.
2.2 Boosting complexity with branched junctions
Three-dimensional DNA structures have great potential for
encapsulating and delivering drugs, proteins and nanomaterials.
Their confined spaces allow the exploration of new reactivity
and catalysis for guest molecules, and can serve as templates to
grow materials. Using the single-stranded DNA polygons
described above, we showed the construction of a library of
DNA polyhedra, including triangular prisms, cubes, pentagonal
and hexagonal prisms, heteroprisms and biprisms (Fig. 4a).
13
This was done by connecting the DNA polygons top and
bottom, using linking strands, in quantitative yields. We also
showed how these DNA cages can be structurally expanded
and contracted between three different sizes, using externally
added DNA strands. These capsules are the size of proteins,
yet their structure can be varied at will and they are fully
addressable and switchable. Recently, we showed a simple
method to build these 3D-DNA structures from a minimum
number of commercially available DNA strands, in quantitative
yields at room temperature.
14
This method is readily accessible
to any laboratory, even without prior training in DNA chemistry.
We then reported a modular approach for the synthesis of
DNA nanotubes, by assembling these DNA polygons long-
itudinally.
15
This method gave ready control of the geometry,
size and stiffness of these assemblies, and ability to generate
these in single-stranded and ‘open’, and double-stranded and
‘closed’ forms. We demonstrated their ability to encapsulate
and selectively release cargo, by creating a DNA nanotube
that encapsulates gold nanoparticles into ‘pea-pod’ lines
(Fig. 4b).
16
When a specific DNA strand is added, it removes
the strands that close the nanotube, and the nanoparticle
cargo is released. We also reported a method to precisely
control the length of DNA nanotubes, and allow them to be
modified independently at every one of their layers.
17
These
materials are promising for drug delivery, nanowire growth, as
interconnects and tracks for molecular motors.
Shchepinov,
18
von Kiedrowski
8
and Sawai
19
have reported
three-way organic junctions on which DNA can grow by
standard phosphoramidite chemistry. If self-complementary
strands are used, the resulting junctions self-assemble into wire-
frame cages, analogous to the cryptands of classic host–guest
chemistry. These cages are more thermodynamically stable than
the corresponding unbranched duplexes, suggesting a highly
cooperative assembly due to the vertex effectively preorganizing
the linking strands. Again here, self-complementarity results in
a concentration-dependent product distribution rather than a
single assembly. Tris-branched motifs using a 1,3,5-tri-substituted
benzene vertex also allows unique sequences to be addressed
on each arm of the synthetic insertion.
20
Such insertions led to
a small hexagonal nanostructure that could be modified with
FRET (fluorescence resonance energy transfer) pairs at pre-
determined positions along the DNA scaffold. More recently,
this same modification was used to create an asymetric four-
ring network assembled in a one-step annealing process, which
can be covalently ligated using site-specific click chemistry.
21
von Kiedrowski and coworkers have used a benzene scaffold
to create tris-oligonucleotide branched linkers with C
3h
sym-
metry that could be used to prepare a 3D DNA dodecahedron
(Fig. 5a).
22
Bao and coworkers further explored organic molecule-DNA
hybrid structures by looking at mono-, bis- and tris-branched
Fig. 4 Dynamic DNA cycles allow the preparation of (a) 3D prisms
(Adapted with permission from ref. 13. Copyright 2007 American
Chemical Society) and (b) nanotubes that can be used for nanoscale
organization.
Fig. 5 Branched DNA systems. (a) A DNA dodecahedron prepared
from a tris-branched junction. Reproduced from ref. 22 with permis-
sion from Wiley-VCH. (b) Rolled up tiles can be prepared using a
porphyrin-DNA four-way junction. Reproduced from ref. 24 with
permission from Wiley-VCH.
This journal is cThe Royal Society of Chemistry 2011 Chem. Soc. Rev., 2011, 40, 5647–5656 5651
aromatic compounds that do not require traditional phos-
phoramidite chemistry.
23
Cross-coupling routes using iso-
thiocyanate-‘click’ and amide coupling strategies produced
hybrid structures containing polyethylene glycol (PEG),
poly(p-phenylene ethynylene) (PPE), and benzene-tricarboxylate
and single-stranded DNA that could then be enzymatically
elongated to prepare micrometre-sized double-stranded systems.
Endo, Majima, Seeman
24
and Shchepinov
18
have succeeded
in creating functional higher-order branching junctions as well.
Rather than discrete cages, these junctions template extended
systems with significant cross-linking. But using careful design,
assembly can be controlled to yield well-defined structures.
Fig. 5b illustrates an elegant example of this, where a porphyrin-
based four-way junction is combined with a DNA tile system
known to form 2D arrays.
24
One of the tiles was tagged with
an extra strand, complementary to the junction branches. This
tagging preorganized the tile into circular tetramers, ensuring
that subsequent tile association led to rolled up fibers rather
than the flat sheets normally observed.
Extending the synthetic core to a tetra-substituted building
block produces a DNA hybrid structure with an additional
element of self-assembly. In an attempt to examine the possibility
of DNA crystallization, Richert and coworkers pre-oriented
DNA of various sequence lengths onto a tetrahedral core.
25
Through a two base-pair interaction on each arm, a cross-linked
material assembled in a stable and selective way (Fig. 6a). It
should be noted that a hallmark in structural DNA nano-
technology was recently achieved with Seeman and coworkers
preparing the first self-assembled DNA crystals.
26
Although the
DNA crystals were made without synthetic modification, they
exhibited similar properties to the tetrahedral DNA hybrid and
remarkable stability through the two base-pair interaction. These
features support the unique characteristics of 3D DNA materials
and the different rules for their construction. More recently,
synthetic cores have been expanded to six DNA arms with a
pseudo-octahedral structure and an adamantane derivative.
27
Seminal research from Schatz and Nguyen has provided
key insights into some of the nanostructures generated from
bis-
28
and tris-branched
29,30
rigid organic cores (Fig. 6b). The
DNA-hybrid structures formed using these synthetic insertions
display unique cooperativity due to neighboring-duplex inter-
actions and increased effective concentrations. Using a blend
of molecular modeling and experimental techniques, these
researchers are thus generating useful models to help define
some of the cooperative and thermodynamic aspects of DNA-
hybrid materials, thus assisting in future design strategies.
2.3 Incorporating transition metal vertices
While the DNA-synthetic hybrids outlined in previous sections
rely on organic insertions, another important self-assembly
approach takes advantage of transition metal-based environ-
ments.
31
Transition metal centres provide access to a rich
diversity of coordination geometries, bond angles and functional
characteristics unavailable in traditional carbon-based systems.
Nature derives electrostatic stabilization, biomolecular stability/
folding and functional diversity by harnessing the properties
of transition metal centres. By modifiying oligonucleotides
with either metal-complexes as vertices or transition metal
coordinating ligands, researchers can create nanomaterials
with added redox, catalytic, photochemical and synthetic
properties. What’s more, the strong binding displayed by
certain coordination environments can significantly stabilize
DNA materials.
32
A diverse selection of ligand modifications
has allowed researchers to explore the selective coordination
of many different transition metals within DNA. We direct the
reader to recent reviews that show both seminal contributions
and advancements made in this field over the last decade that
specifically pertain to linear DNA-metal construction.
33,34
Early examples of direct metal-complex insertions into DNA,
where the coordination environment influenced the arrange-
ment of DNA strands, were reported by our group
35,36
and
McLaughlin’s.
37,38
In our system, a two-way DNA junction
based on a [Ru(bpy)
3
]
2+
insertion was used to template pre-
dominantly dimeric DNA-metal macrocycles (Fig. 7). A larger
cyclic product distribution was observed when the metal was
absent, leaving a more flexible bipyridine vertex. This illustrates
the role of the metal in directing the assembly process and
favouring one product out of several pre-existing possibilities.
While direct insertion of metal complexes as vertices creates a
wealth of geometric and coordination environments, it should
Fig. 6 (a) Tetra-substituted organic vertices modified with two bases
can be used to create macroscopic materials. Reproduced from ref. 25
with permission from Wiley-VCH. (b) Theoretical and experimental
work with bis DNA-organic vertices illustrates the enhanced coopera-
tivity of nanostructure assembly. Adapted with permission from
ref. 28. Copyright 2010 American Chemical Society.
Fig. 7 Metal complexes can be introduced into DNA to form
branched motifs that assemble into dimeric nanostructures. Repro-
duced from ref. 36 with permission from Wiley-VCH.
5652 Chem. Soc. Rev., 2011, 40, 5647–5656 This journal is cThe Royal Society of Chemistry 2011
be noted that these metals must be sufficiently inert to withstand
automated DNA synthesis conditions.
An alternative strategy is to directly modify DNA with
ligands. Researchers can then use the ligand-metal interaction
or templating ability of DNA to form stable metal coordination
environments within a variety of 2D and 3D nanostructures. A
particularly important example of this strategy involves the
preparation of artificial nucleobases that are capable of metal
binding, and their use to create multi-metal ‘metallobase-paired’
DNA.
33
In this way, problems incorporating metal-complexes
into DNA via solid-phase synthesis can be avoided.
Han and co-workers demonstrated that ssDNA end-modified
with a terpyridine ligand could be used to drive assembly upon
addition of specific metals.
39
When Fe
2+
was added to this
system, the ligand-metal coordination clipped together two
strands of DNA into metal-bound branched molecules
(Fig. 8a). For this process to result in assembly, the two arms
of a given dimer must be unique. A purification step was
therefore necessary in order to separate the heterodimer from
the homodimer side products. This approach gave the
researchers dual modes of control over the assembly process;
both metal coordination and predesigned base-pairing could
be used to define the final product.
A system devised by Kramer and co-workers
40
used a short
DNA sequence doubly end-modified with terpyridine derivatives,
creating a system that would cyclize upon addition of metals
such as Fe
2+
,Zn
2+
and Ni
2+
(Fig. 8b). Here, the thermo-
dynamics of the metal-coordination were strong enough to
displace a complementary DNA strand and allow the system
to cyclize. The stimuli-responsiveness of this structure was in
turn demonstrated by adding excess metal, which opened the
cyclized structure and restored the duplex DNA. Sugimoto
and co-workers have used ligand insertions to assemble higher
order structures. In the presence of Ni
2+
ions, bipyridine-
modified G-quadruplexes change from a anti-parallel to parallel
structure, resulting in intermolecular association to form
‘G-wires’. This type of metal-mediated self-assembly could
be reversed upon addition of EDTA.
41
These contributions
show an impressive level of dynamic control and specificity in
such supramolecular interactions.
Another approach to incorporate transition metal vertices
involves site-specifically inserting ligands into the DNA back-
bone and using DNA self-assembly to generate environments
that bind and stabilize transition metals. An early example by
Gothelf and co-workers showed modular assembly of DNA-
programmed linear and branched conjugated nano-
structures using a metal-salen interaction.
42
Our group selec-
tively incorporated a range of reactive transition metals into
DNA-templated junctions.
32
This involves the creation of an
architecture in which the ligand environment is in close
contact with DNA, allowing the DNA double helix and the
metal complex to synergistically stabilize each other (Fig. 9a).
It results in metallated DNA junctions with unusually high
stability (e.g.,401C melting temperature increase for a 10-base
DNA duplex upon single metal incorporation).
The real potential of DNA lies in its ability to organize
different transition metals in precise locations within a nano
structure. With this in mind, we recently reported the DNA-
templated creation of three different ligand environments.
43
Each of these binds a specific transition metal ion, with
enhancement of the stability of DNA (Fig. 9b). Moreover,
when the ‘incorrect’ metal ion is placed within one of these
ligand environments, ‘error correction’ spontaneously occurs.
Thus, the metal adjusts its oxidation state, displaces a rela-
tively labile metal to form a more stable complex, or causes
reorganization of the coordination site to create the favored
ligand complex.
Using this method, we reported the construction of a metal-
DNA cage, with site-specific incorporation of transition metals
in the vertices of this structure (Fig. 10a).
44
This class of materials
brings together the properties of metal–organic frameworks
(MOF), such as metal-mediated redox, photochemical, magnetic
or catalytic control on encapsulated guest molecules, with the
programmability of DNA and its facile chemical and structural
variation.
One of the consequences of using metal-DNA hybrid systems
is the ability to create ‘DNA-economical’ structures, where the
metal coordination geometry defines the geometry of the
vertex, and the DNA strands give programmable regions that
can used as building blocks for 2D and 3D structures.
Towards this goal, we recently developed a DNA-templated
method to create chiral metal-DNA junctions.
45
By modifying
DNA strands with a phenanthroline ligand and templating
using a minimal amount of DNA, we generated a chiral DNA
Fig. 8 Ligand modified DNA and nanostructure assemble via ligand-
metal interactions. (a) Terpyridine modified DNA drives the for-
mation of a DNA triangle using Fe
2+
coordination. Adapted with
permission from ref. 39. Copyright 2004 American Chemical Society.
(b) Zn
2+
drives the opening and closing of DNA cycles. Adapted with
permission from ref. 40. Copyright 2005 American Chemical Society.
Fig. 9 Metal DNA complexes and structures. (a) Site-specifically
incoporating diphenylphenanthroline (dpp) into DNA templates
nanostructures with highly stable duplexes. Reproduced from ref. 32
with permission from Wiley-VCH. (b) Using two different ligand
insertions, selective coordination environments can be templated with
DNA. Reproduced from ref. 43 with permission from Wiley-VCH.
This journal is cThe Royal Society of Chemistry 2011 Chem. Soc. Rev., 2011, 40, 5647–5656 5653
junction with a transition metal at the branch point (Fig. 10b).
This junction contains four different single-stranded DNA
arms, making it a unique synthon for further self-assembly.
Its information content and precise spatial arrangement of
DNA were then used to construct metal-DNA triangular
rungs that could be assembled into the first metal-DNA
nanotubular structure.
The functions of such multi-metallic systems have yet to be
studied in detail, but these studies have shown that transition
metals can be selectively incorporated to generate monodisperse
and highly stabilized DNA structures with preserved self-
assembly properties, smaller sizes and dynamic character.
Future applications envisaged for metal-containing DNA
materials include catalysis, sensing, nanoelectronics and artificial
photosynthesis systems.
3. Self-assembly driven by chemical modification
So far, our discussion has focused on hybrid DNA systems,
where assembly properties are still primarily determined by
DNA Watson–Crick base pairing. Synthetic insertions have
the additional potential to introduce new assembly features
into DNA, that complement or possibly even override DNA’s
own base-pairing. Because of the wealth of supramolecular
motifs that have been achieved in the last four decades, this
nascent area holds great potential towards expanding DNA
nanotechnology into completely new structures.
3.1 Using molecules to drive DNA self-assembly
A number of research groups have used DNA as a scaffold to
organize polyaromatic molecules into p-stacked arrays. This
was achieved either by replacing the natural DNA bases with
artificial polyaromatic molecules, or by site-specifically attaching
aromatic chromophores to these bases. Unique photophysical
properties, unusual stabilities and biodetection applications
can result from this mode of DNA modification.
46,47
Hydrogen
bonding has also been used to reprogram DNA hybridization
into a five-stranded structure by replacing the nucleobases
with iso-guanine, resulting in pentameric assemblies.
48
We
again direct the reader to a number of recent reviews that
outline both the seminal contributions and advancements
made in this area.
Perylene diimide-oliugonucleotides have been shown to
form either duplex or hairpin structures based on the choice
of DNA sequences used.
49
Wagenknecht has used Y-shaped
DNA junctions end-terminated with perylene units to drive
self-assembly in 2D.
50
Extended aromatic insertions of this
type can produce specific interactions with other hydrophobic
materials. As an example, porphyrin modifications can anchor
a DNA hexagon to a lipid membrane,
51
yielding a dynamic yet
surface-bound assembly (Fig. 11a). The weak anchoring allows
DNA components to migrate and adjust to their environment
without leaving the membrane surface. This could serve as an
important step towards self-repairing or stimuli-responsive
materials.
DNA foldamers are an interesting example of the effect of
synthetic insertions on self-assembly. A foldamer is a polymer
that wraps into a highly specific conformation under given
conditions, using a number of non-covalent interactions
between its monomer units. Synthetic insertion of hydrophobic
perylene units within a hydrophilic DNA chain can drive
folding processes, to yield a final product that is different
from the standard double helix.
52
This folded structure places
the perylene units into a p-stacked aromatic core. This behaviour
was even more favourable with increasing temperature (a condi-
tion that normally denatures secondary and tertiary struc-
tures), due to endothermic folding enthalpies. With careful
sequence planning, the group designed a star-shaped foldamer
that is stable across a wide temperature range; at low tempera-
tures DNA base-pairing maintains the structure, while at high
temperatures hydrophobic effects fill this role (Fig. 11b). Nucleic
acids are known to carry out some catalytic functions, which
means that highly stable hybrids with well-defined folding have
exciting implications for mimicking folded, functional proteins.
3.2 DNA-block copolymers and higher order assembly
To increase the long-range order of assembled structures,
researchers are working to covalently attach biomolecules to
polymers. The resulting products show new biological and self
assembly properties and are being explored for a variety of
applications, from therapeutics and drug delivery, to tissue
engineering and materials science.
53–55
In the context of
supramolecular DNA assembly, DNA-lipid and block copolymer
conjugates have been prepared and can assemble into a
number of interesting morphologies.
56
The amphiphilic nature
Fig. 10 Metal DNA complexes and structures in 3D. (a) Metal-DNA
cage structure with selective coordination environments templated in
3D. (b) Chiral four-way metal-DNA junction formed via templated
assembly of modified sequences. Reproduced from ref. 45 with
permission from Wiley-VCH.
Fig. 11 Self-assembly via chemical modification (a) Porphyrin-DNA
modifications can be used to selectively attach DNA nanostructures to
lipid vesicles. Reproduced from ref. 51 with permission from Wiley-
VCH. (b) DNA foldamers yield folded structures through p-stacking
interactions. Adapted with permission from ref. 52. Copyright 2003
American Chemical Society.
5654 Chem. Soc. Rev., 2011, 40, 5647–5656 This journal is cThe Royal Society of Chemistry 2011
of these materials has yielded interesting micellar assemblies
that can be further manipulated with the recognition properties
of DNA. We direct the reader to recent reviews that outline
some of the properties of biohybrid materials prepared by
attaching lipids or polymers to DNA.
53,56,57
Switchable self-assembled structures were created using
polypropyleneoxide (PPO) and polyethylene glycol (PEG)
block copolymers with short DNA tails.
58
Alone, these struc-
tures formed star micelles with hydrophobic cores and oligo-
nucleotide coronas. But when long template strands composed
of repeating complement sequences to the DNA block are
added, a new arrangement was formed. Elongated assemblies
of duplex DNA were linked by repeating hydrophobic cores,
as shown in Fig. 12a, with size defined by the choice of
template strand. It is important to note that adding shorter
complementary sequences did not disrupt the micelle in this
design, making such structures amenable to functionalization.
The organized DNA within these micelle structures has also
been used to template the assembly of the Cowpea Chlorotic
Mottle Virus (CCMV) capsid, which traps or encapsulates
hydrophobic and hydrophilic molecules in the core of the
particle.
59
Dynamic character was shown by Gianneshi and
co-workers, who designed a PEG-DNA-brush block copolymer
with the ability to shift between spherical and cylindrical
shaped micelles with the addition of complementary DNA
strands or DNAzymes (Fig. 12b).
60
Our group employed the self-assembling properties of short
oligoethylene glycol (OEG) units arranged in a dendritic
fashion.
61
By modifying the ends of DNA strands with these
dendritic structures and hybridizing the strands into a duplex,
a block-copolymer architecture was formed (Fig. 13). These
hybrid structures, composed of a dsDNA core surrounded by
the OEG dendritic structure, self-assembled into long-range
fibers in organic solvents by end to end stacking of the
DNA strands. Through simple modification of the deposition
conditions, the fibers could also assemble into long-range
ordered 2D networks. The Liu group created an amphiphilic
DNA-dendron hybrid, containing a hydrophobic dendron
attached to a DNA single strand. In aqueous media, this
molecule assembles into nanofibers, containing a hydrophobic
dendron core and a single-stranded DNA corona, and these
morphologies were shown to encapsulate hydrophobic guest
molecules.
62
Recently, the groups of Tan
63
and Mirkin
64
have
demonstrated that self-assembled polyvalent DNA structures
display efficient cellular uptake and thus may prove to be
candidates for drug delivery applications.
By tuning the hydrophilic and hydrophobic properties of
the organic insertion, DNA conjugates can be designed to
exhibit unique properties that are not possible in the individual
components themselves. The result is a supramolecular system
that can display dynamic and thermoresponsive characteristics
as well as the ability to achieve long-range ordering of materials
at the nanometre length-scale in a well-defined manner.
3.3 Dynamic control over DNA self-assembly
Our description of supramolecular DNA assembly has thus far
focused on covalent modification with molecules that impart
structural or self-assembling properties to DNA nanostructures.
In this final section, we shift the discussion to molecules that
can dynamically control DNA assembly, either through covalent
insertion or non-covalent interactions. In DNA nanomachines,
control over DNA motion occurs in the presence of externally
added DNA strands, enzymes, or in response to pH or ion
changes.
65
The use of synthetic molecules to effect DNA
structural switching or modification can significantly widen
the spectrum of dynamic control available for further tuning
the assembly of 2D and 3D nanostructures.
One of the ways to control DNA assembly is by developing
molecules that ‘repair’ DNA mismatches and restore DNA
hybridization. Nakatani and co-workers introduced a small
synthetic ligand, a naphthyridinecarbamate dimer (NCD), that
acts as a‘molecular glue’ in DNA nanofabrication.
66
Recently,
this molecule was used to modulate DNA hybridization and
Fig. 12 DNA-block copolymer conjugates. (a) Micelles assemble
using amphiphiles made of DNA and block copolymers. Reproduced
from ref. 58 with permission from Wiley-VCH. (b) Using specific
DNA inputs, a PEG-DNA-brush polymer can shift shape between
spherical and cylindrical micelles. Reproduced from ref. 60 with
permission from Wiley-VCH.
Fig. 13 Covalently modifying DNA with dendritic oligoethylene
moieties enables self-assembly of long-range fibers. Adapted with
permission from ref. 61. Copyright 2010 American Chemical Society.
Fig. 14 Molecule mediated DNA self-assembly. (a) Small molecules
can selectively interact with GG mismatches sites, allowing for open-
ing/closing of 3D DNA nanostructures. Reproduced from ref. 67 with
permission from the Royal Society of Chemistry. (b) Incorporating
azobenzenes into a 3D DNA tetrahedron allows reversible cycling of
structure. Reproduced from ref. 72 with permission from the Royal
Society of Chemistry.
This journal is cThe Royal Society of Chemistry 2011 Chem. Soc. Rev., 2011, 40, 5647–5656 5655
form a DNA tetrahedron (Fig. 14a).
67
This method requires
that key structural sequences involved in nanostructure for-
mation be altered with GG mismatches. The set of component
DNA strands are designed to assemble into a tetrahedral struc-
ture only in the presence of the small molecule NCD, which
selectively binds at the GG-mismatch site. Upon coordination,
NCD increases the thermal stability of dsDNA and allows 3D
assembly to take place. Another method to control DNA
assembly is to specifically block or disrupt the hydrogen
bonding interactions required for strands to associate. A
photocaged thymidine residue has been incorporated into
DNA by standard synthesis conditions, disrupting hydrogen
bonding.
68
In this site-specific, covalent modification strategy,
DNA hybridization was restored by irradiating to ‘decage’
thymine’s hydrogen bonding face.
Within the supramolecular toolbox, molecules that undergo
photoswitchable behaviour provide impressive dynamic control
over self-assembly properties.
69
The thermodynamically
favoured association of the DNA duplex, for example, can
be overridden by altering the conformation of molecules that
interact with it. Some of the most commonly used molecules
are azobenzene and its derivatives, where photo-isomerization
between cis and trans forms provides enough motion to
destabilize duplex formation.
70
Dynamic control over two-
dimensional systems, such as a DNA hairpin, has been
achieved, revealing that the correct placement and number
of azobenzene derivatives is paramount to initiating high
levels of assembly control.
71,72
Tan and co-workers recently
switched 3D DNA assembly with an azobenzene derivative.
72
In this example, the shape of a DNA tetrahedron can be
reversibly controlled by alternately irradiating with different
wavelengths of light (Fig. 14b).
An alternative approach to controlling 2D and 3D DNA
assembly is to use molecules that can selectively template how
specific nanostructures are formed from a dynamic equili-
brium of constituents. This method is based on the highly
successful area of dynamic combinatorial chemistry.
3,73
It can
result in correcting assembly errors and reducing the number
of strands required to prepare complex DNA nanostructures.
Our group demonstrated this strategy by constructing 5,50-
branched DNA building blocks, containing a rigid organic
vertex and two identical DNA arms. In the absence of
template, the assembly of one of these molecules with its
complementary partner produces a large number of cyclic
nanostructures, from dimer to square to hexamer to oligomers.
When the molecule ruthenium tris-bipyrdine (Ru(bpy)
32+
)is
added, the entire library converges into a square product in
quantitative yields. While further mechanistic investigation is
underway, this example illustrates the ability of externally
added molecules to dramatically affect DNA self-assembly
and result in clean formation of a single product from a small
number of symmetrical DNA strands (Fig. 15).
74
Small molecules that can change the properties of DNA
hybridization are a powerful means of generating 2D and 3D
nanostructures with dynamic character. These can be applied
as ‘error-correcting’ tools for DNA self-assembly and as
stimuli-responsive systems for drug delivery and biosensing.
4. Conclusions and outlook
In this review we have outlined examples where the synergistic
arrangement of nucleic acids and synthetic components can be
used to address challenges of precise molecular patterning and
chemical construction. Interweaving the superior self-assembly
properties of DNA with the functional elements of synthetic
organic and inorganic molecules has, to date, created nano-
structures that display properties far greater than the sum of
their parts. So where does the field of supramolecular DNA
assembly go from here? Research aimed at developing a
more fundamental understanding of the influence of synthetic
insertions on DNA structure is now necessary. In addition,
increasing interest in biology and medicine will drive discovery
into functional DNA-based nanostructures. This will make it
ever more important to develop ways to make DNA-based
materials compatible and stable under biological conditions.
In principle, synthetic insertions can result in ‘DNA-minimal’
hybrid materials. In these structures, DNA strands act as a
‘frame’ to provide programmability and structural definition.
These DNA frames would be used to guide the organization
of synthetic materials that are engineered for increased biocompat-
ibility, cell penetration ability, and decreased toxicity. Synthetic
modifications will allow one to be more conservative in the use of
nucleic acid components, while maximizing their structural and
functional roles. Towards mastering both complexity and function,
continued effort on the part of researchers to create geometrically
well-defined 2D and 3D architectures from synthetically modified
nucleic acids and assess their properties will be critical.
Acknowledgements
The authors thank NSERC, CFI, CSACS, QNRF and CIFAR
for financial support, and Alison Palmer for editorial help on
the manuscript. H.F.S is a Cottrell Scholar of the Research
Corporation.
References
1 N. C. Seeman, Nature, 2003, 421, 427–431.
2 C. Lin, Y. Liu and H. Yan, Biochemistry, 2009, 48, 1663–1674.
3 J.-M. Lehn, Chem. Soc. Rev., 2007, 36, 151–160 and references
therein.
Fig. 15 Small molecule ruthenium tris-bypyridine (Ru(bpy)
32+
)is
used to selectively control the 2D DNA nanostructure formation of
branched symmetric starting materials. Adapted with permission from
ref. 74. Copyright 2007 American Chemical Society.
5656 Chem. Soc. Rev., 2011, 40, 5647–5656 This journal is cThe Royal Society of Chemistry 2011
4 M. D. Pluth and K. N. Raymond, Chem. Soc. Rev., 2007, 36,
161–171 and references therein.
5 S. J. Loeb, Chem. Soc. Rev., 2007, 36, 226–235 and references
therein.
6 F. A. Aldaye, A. L. Palmer and H. F. Sleiman, Science, 2008, 321,
1795–1799.
7 J. Shi and D. E. Bergstrom, Angew. Chem., Int. Ed. Engl., 1997, 36,
111–113.
8 M. Scheffler, A. Dorenbeck, S. Jordan, M. Wu
¨stefeld and G. von
Kiedrowski, Angew. Chem., Int. Ed., 1999, 38, 3312–3315.
9 F. A. Aldaye and H. F. Sleiman, Angew. Chem., Int. Ed., 2006, 45,
2204–2209.
10 F. A. Aldaye and H. F. Sleiman, J. Am. Chem. Soc., 2007, 129,
4130–4131.
11 D. Y. Zhang and G. Seelig, Nat. Chem., 2011, 3, 103–113.
12 T. Zhang, Z. Yang and D. Liu, Nanoscale, 2011, 3, 4015–4021 and
references therein.
13 F. A. Aldaye and H. F. Sleiman, J. Am. Chem. Soc., 2007, 129,
13376–13377.
14 C. K. McLaughlin, G. D. Hamblin, F. A. Aldaye, H. Yang and
H. F. Sleiman, Chem. Commun., 2011, 47, 8925–8927.
15 F. A. Aldaye, P. K. Lo, P. Karam, C. K. McLaughlin, G. Cosa and
H. F. Sleiman, Nat. Nanotechnol., 2009, 4, 349–352.
16 P. K. Lo, P. Karam, F. A. Aldaye, C. K. McLaughlin,
G. D. Hamblin, G. Cosa and H. F. Sleiman, Nat. Chem., 2010,
2, 319–328.
17 P. K. Lo, F. Altvater and H. F. Sleiman, J. Am. Chem. Soc., 2010,
132, 10212–10214.
18 M. S. Shchepinov, K. U. Mir, J. K. Elder, M. D. Frank-
Kamenetskii and E. M. Southern, Nucleic Acids Res., 1999, 27,
3035–3041.
19 T. Kuroda, Y. Sakaurai, Y. Suzuki, A. O. Nakamura,
M. Kuwahara, H. Ozaki and H. Sawai, Chem.–Asian J., 2006, 1,
575–580.
20 J. Tumpane, et al.,Chem. Phys. Lett., 2007, 440, 125–129.
21 E. P. Lundberg, C. Plesa, L. M. Wilhelmsson, P. Lincoln,
T. Brown and B. Norden, ACS Nano, 2011, 5, 7565–7575.
22 J. Zimmermann, M. P. J. Cebulla, S. Mo
¨nninghoff and G. von
Kiedrowski, Angew. Chem., Int. Ed., 2008, 47, 3626–3630.
23 J. K. Lee, Y. H. Jung, J. B.-H. Tok and Z. Bao, ACS Nano, 2011,
5, 2067–2074.
24 M. Endo, N. C. Seeman and T. Majima, Angew. Chem., Int. Ed.,
2005, 44, 6074–6077.
25 M. Meng, et al.,ChemBioChem, 2009, 10, 1335–1339.
26 J. Zheng, J. J. Birktoft, Y. Chen, T. Wang, R. Sha,
P. E. Constantinou, S. L. Ginnell, C. Mao and N. C. Seeman,
Nature, 2009, 461, 74–77.
27 A. Sing, et al.,Angew. Chem., Int. Ed., 2011, 50, 3227–3231.
28 B. R. Stepp, J. M. Gibbs-Davis, D. L. F. Koh and S. T. Nguyen,
J. Am. Chem. Soc., 2008, 130, 9628–9629.
29 J. T. R. Prytkova, I. Eryazici, B. Stepp, S.-B. T. Nguyen and
G. C. Schatz, J. Phys. Chem. B, 2010, 114, 2627–2634.
30 K. I. Eryazici, T. R. Prytkova, G. C. Schatz and S.-B. T. Nguyen,
J. Am. Chem. Soc., 2010, 132, 17068–17070.
31 S. Leininger, B. Olenyuk and P. J. Stang, Chem. Rev., 2000, 100,
853–907.
32 H. Yang and H. F. Sleiman, Angew. Chem., Int. Ed., 2008, 47,
2443–2446.
33 G. H. Clever and M. Shionoya, Coord. Chem. Rev., 2010, 254,
2391–2402.
34 H. Yang, K. L. Metera and H. F. Sleiman, Coord. Chem. Rev.,
2010, 254, 2403–2415.
35 I. Vargas-Baca, D. Mitra, H. Zulyniak, J. Banerjee and
H. F. Sleiman, Angew. Chem., Int. Ed., 2001, 40, 4629–4632.
36 D. Mitra, N. Di Cesare and H. F. Sleiman, Angew. Chem., Int. Ed.,
2004, 43, 5804–5808.
37 K. M. Stewart and L. W. McLaughlin, J. Am. Chem. Soc., 2004,
126, 2050–2057.
38 K. M. Stewart, J. Rojo and L. W. McLaughlin, Angew. Chem., Int.
Ed., 2004, 43, 5808–5811.
39 J. S. Choi, C. W. Kang, K. Jung, J. W. Yang, Y.-G. Kim and
H. Han, J. Am. Chem. Soc., 2004, 126, 8606–8607.
40 M. Go
¨ritz and R. Kra
¨mer, J. Am. Chem. Soc., 2005, 127,
18016–18017.
41 D. Miyoshi, H. Karimata, Z.-M. Wang, K. Koumoto and
N. Sugimoto, J. Am. Chem. Soc., 2007, 129, 5919–5920.
42 K. V. Gothelf, A. H. Thomsen, M. Nielsen, E. Clo and
R. S. Brown, J. Am. Chem. Soc., 2004, 126, 1044–1046.
43 H. Yang, A. Rys, C. K. McLaughlin and H. F. Sleiman, Angew.
Chem., Int. Ed., 2009, 48, 9919–9923.
44 H. Yang, et al.,Nat. Chem., 2009, 1, 390–396.
45 H. Yang, F. Altvater, A. D. de Bruijn, C. K. McLaughlin, P. K. Lo
and H. F. Sleiman, Angew. Chem. Int. Ed., 2011, 50, 4620–4623.
46 K. T. J. Bandy, A. Brewer, J. R. Burns, G. Marth, T. Nguyen and
E. Stulz, Chem. Soc. Rev., 2011, 40, 138–148 and references therein.
47 A. Ruiz-Carretero, P. G. A. Janssen, A. Kaeser and A. P. H.
J. Schenning, Chem. Commun., 2011, 47, 4340–4347 and references
therein.
48 J. C. Chaput and C. Switzer, Proc. Natl. Acad. Sci. U. S. A., 1999,
96, 10614–10619.
49 Z. Zheng, H. Long, G. C. Schatz and F. D. Lewis, Chem.Commun.,
2005, 4795–4797.
50 F. Menacher, V. Stepanenko, F. Wu
¨rthner and
H.-A. Wagenknecht, Chem.–Eur. J., 2011, 17, 6683–6688.
51 K. Bo
¨rjesson, E. P. Lundberg, J. G. Woller, B. Norde
´n and
B. Albinsson, Angew. Chem. Int. Ed., 2011, 50, 8312–8315.
52 W. Wang, W. Wan, H.-H. Zhou, S. Niu and A. D. Q. Li, J. Am.
Chem. Soc., 2003, 125, 5248–5249.
53 G. Fuks, R. M. Talom and F. Gaufree, Chem. Soc. Rev., 2011, 40,
2475–2493.
54 M. J. Webber, J. A. Kessler and S. I. Stupp, J. Intern. Med., 2009,
267, 71–88.
55 J. M. Holzwarth and P. X. Ma, J. Mater. Chem., 2011, 21,
10243–10251.
56 M. Kwak and A. Herrmann, Angew. Chem., Int. Ed., 2010, 49,
8574–8587.
57 A. Patwa, A. Gissot, I. Bestel and P. Barthelemy, Chem. Soc. Rev.,
2011, DOI: 10.1039/c1cs15038c.
58 K. Ding, F. E. Alemdaroglu, M. Bo
¨rsch, R. Berger and
A. Herrmann, Angew. Chem., Int. Ed., 2007, 46, 1172–1175.
59 M. Kwak, I. J. Minten, D.-M. Anaya, A. J. Musser, M. Brasch, R.
J. M. Nolte, K. Mu
¨llen, J. J. L. M. Cornelissen and A. Herrmann,
J. Am. Chem. Soc., 2010, 132, 7834–7835.
60 M.-P. Chien, A. M. Rush, M. P. Thompson and N. C. Gianneshi,
Angew. Chem., Int. Ed., 2010, 49, 5076–5080.
61 K. M. M. Carneiro, F. A. Aldaye and H. F. Sleiman, J. Am. Chem.
Soc., 2010, 132, 679–685.
62 L. Wang, et al.,Soft Matter, 2011, 7, 7197–7190.
63 Y. Wu, K. Sefah, H. Liu, R. Wang and W. Tan, Proc. Natl. Acad.
Sci. U. S. A., 2010, 107, 5–10.
64 J. I. Cutler, K. Zhang, D. Zheng, E. Auyeung, A. E. Prigodich and
C. A. Mirkin, J. Am. Chem. Soc., 2011, 133, 9254–9257.
65 Y. Krishnan and F. C. Simmel, Angew. Chem., Int. Ed., 2011, 50,
3124–3156.
66 C. Dohno and K. Nakatani, Chem. Soc. Rev., 2011, DOI: 10.1039/
c1cs15062f.
67 C. Dohno, H. Atsumi and K. Nakatani, Chem. Commun., 2011, 47,
3499–3501.
68 H. Lusic, D. D. Young, M. O. Lively and A. Deiters, Org. Lett.,
2007, 9, 1903–1906.
69 G. Mayer and A. Heckel, Angew. Chem., Int. Ed., 2006, 45, 4900–4921.
70 A. A. Beharry and G. A. Woolley, Chem. Soc. Rev., 2011, 40,
4422–4437.
71 X. Liang, T. Mochizuki and H. Asanuma, Small, 2009, 5,
1761–1768.
72 D. Han, J. Huang, Z. Zhu, Q. Yuan, M. You, Y. Chenab and
W. Tan, Chem. Commun., 2011, 47, 4760–4672.
73 P. T. Corbett, et al.,Chem. Rev., 2006, 106, 3652–3711.
74 F. A. Aldaye and H. F. Sleiman, J. Am. Chem. Soc., 2007, 129,
10070–10071.
