Self-assembled multivalent RGD-peptide arrays-morphological control and integrin binding

Article abstract of DOI:10.1039/c3ob00034f

We report the synthesis of four different RGD peptide derivatives which spontaneously self-assemble into nanoscale architectures. Depending on the information programmed into the molecular-scale building blocks by organic synthesis, these compounds assemble into different nanoscale morphologies. This process can be fully understood using multiscale modelling which provides predictive insight into subtle differences, such as whether the compounds form spherical micelles, rod-like cylinders or tubular assemblies, and predicts experimentally observed critical aggregation concentrations (CACs). We then probe the multivalent binding of these assemblies to integrin proteins and demonstrate that the spherical micellar assemblies perform well in our solution-phase integrin binding assay as a consequence of self-assembled multivalency, with the CAC switching-on the binding. Conversely, the cylindrical assemblies do not work in this assay. As such, the nanoscale morphology controls the apparent ability to perform as a self-assembled multivalent ligand array. The Royal Society of Chemistry 2013.

Full text of DOI:10.1039/c3ob00034f

Organic &
Biomolecular Chemistry
View Article Online
View Journal
PAPER
Self-assembled multivalent RGD-peptide arrays –
morphological control and integrin binding†
Cite this: DOI: 10.1039/c3ob00034f
Daniel J. Welsh,a Paola Posocco,b Sabrina Pricl*b,c and David K. Smith*a
We report the synthesis of four different RGD peptide derivatives which spontaneously self-assemble into
nanoscale architectures. Depending on the information programmed into the molecular-scale building
blocks by organic synthesis, these compounds assemble into different nanoscale morphologies. This
process can be fully understood using multiscale modelling which provides predictive insight into subtle
differences, such as whether the compounds form spherical micelles, rod-like cylinders or tubular assem-
blies, and predicts experimentally observed critical aggregation concentrations (CACs). We then probe
the multivalent binding of these assemblies to integrin proteins and demonstrate that the spherical
micellar assemblies perform well in our solution-phase integrin binding assay as a consequence of self-
assembled multivalency, with the CAC switching-on the binding. Conversely, the cylindrical assemblies do
not work in this assay. As such, the nanoscale morphology controls the apparent ability to perform as a
self-assembled multivalent ligand array.
Received 8th January 2013,
Accepted 22nd March 2013
DOI: 10.1039/c3ob00034f
www.rsc.org/obc
lectins and hence interfere with viral adhesion.4 Multivalent
cationic arrays can bind polyanions such as DNA5 and
Introduction
Multivalent binding is a powerful strategy to achieve high- heparin,6 with applications in gene delivery and post-surgical
affinity molecular recognition between nanoscale surfaces.1 In therapy. In this way, self-assembled multivalency (SAMul) is a
recent times, there has been increasing attention on the use of viable approach for the generation of new synthetic solutions
self-assembly methods to generate nanoscale multivalent to nanomedicine.7
arrays.2 Self-assembly is a synthetically straightforward strat-
Given the importance of integrin proteins on cell surfaces,8
egy, in which monovalent building blocks spontaneously form there has been intense interest in studying the interaction
higher order architectures.3 Using self-assembly to generate between RGD peptides and peptidomimetics with this class of
multivalent ligand arrays offers a number of advantages over protein.9 Such interactions are vital in mediating the ways in
the covalent construction of such systems: (i) spontaneous which cells interact with the extracellular matrix, and have
simple assembly, (ii) well-defined low-molecular-weight build- therefore been employed in materials with potential appli-
ing blocks suitable for clinical approval, (iii) easily tunable cations in tissue engineering.10 Furthermore, the optimisation
ligands, (iv) tunable nanostructure morphologies, (v) ability to of such interactions has been proposed as a way of targeting
assemble different active components into a single nanostruc- tumour cells which have up-regulated levels of integrin pro-
ture, (vi) simple or triggered disassembly/degradation. By teins.11 On initial inspection, integrins may appear to be
tuning the choice of ligands it is possible to modify these self- unpromising targets for multivalent binding, as they only
assembled multivalent systems to bind a range of different bio- possess a single ligand binding site. However, in biological
logical targets, for example multivalent sugar arrays can bind systems, integrins are often clustered in cell membranes. To
enhance the binding of RGD peptides to integrins, there has
therefore been interest in using multivalent peptide arrays.12
Self-assembled arrays of RGD peptides have been reported in
aDepartment of Chemistry, University of York, Heslington, York, YO10 5DD, UK.
E-mail: david.smith@york.ac.uk; Fax: +44 (0)1904 324156
the literature. RGD-containing lipopeptides have been demon-
bMolecular Simulations Engineering (MOSE) Laboratory, Department of Engineering
strated can be insert into membranes and liposomes which
and Architecture (DEA), University of Trieste, Piazzale Europa 1, 34127 Trieste, Italy
cNational Interuniversity Consortium for Material Science and Technology (INSTM),
gain enhanced affinity for integrins and/or show modified bio-
logical properties.13 Self-assembling RGD-functionalised lipids
have also been shown to assemble into morphologies such as
nanoribbons or fibres.14 Given the importance of RGD–integ-
rin interactions, we have recently been interested in gaining a
fundamental insight into the nature of multivalency in RGD
Research Unit MOSE-DEA, University of Trieste, Italy.
E-mail: Sabrina.Pricl@di3.units.it
†Electronic supplementary information (ESI) available: Synthesis and character-
isation data for all compounds, data from the Nile Red assay for determination
of CAC, data from pyrene fluorescence study, full details of multiscale modelling
approach. See DOI: 10.1039/c3ob00034f
This journal is © The Royal Society of Chemistry 2013
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
Scheme 1 Schematic of self-assembly of RGD-derivatives in aqueous solution to yield a self-assembled multivalent array of RGD peptides.
peptide–integrin interactions and demonstrated that self-
Compound C12-RGD, synthesised by TBTU-mediated coup-
assembled arrays of RGD peptides act in a multivalent manner ling of lauric acid to a protected RGD peptide, followed by
(Scheme 1) analogous to the way in which dendritic molecules global deprotection, was reported by us previously.15 Using the
can arrange multiple ligands, enhancing the interaction with same general synthetic methodology, we coupled behenic acid
integrin proteins within vesicle membranes.15
or 1-pyrenebutyric acid to the same peptide, and generated
In this new paper we demonstrate (i) how the self- C22-RGD or Py-RGD in very good yields of 63% and 77%
assembled approach to multivalency can easily yield a range of respectively (see ESI† for details). We also designed a larger
different nanoscale multivalent structures, (ii) how this precise twin-tailed hydrophobic unit for conjugation to the RGD
morphological control can modify interactions with the bio- peptide to give (C12)2Lys-RGD. This compound was syn-
logical target. Furthermore, using multiscale modelling we thesised according to the method outlined in Scheme 2. The
gain an understanding of subtle effects in the self-assembly hydrophobic unit was synthesised by TBTU-mediated conju-
process. This exemplifies how a combined experimental and gation of lauric acid to lysine methyl ester to provide com-
theoretical approach to self-assembled multivalency will be a pound 1-ester. Basic saponification unmasked the carboxylic
powerful way of designing and developing novel morphologi- acid to give 1-acid which was then coupled with methyl 6-
cally tunable nanoscale systems capable of intervening in bio- aminohexanoate, again by TBTU-mediated peptide coupling
medical processes.
methods to give compound 2-ester. Saponification of the
methyl ester provided 2-acid. The overall yield for this four-
step synthesis was a pleasing 46%. We then needed to attach a
hydrophilic tri-ethylene glycol unit to the RGD peptide in
order to provide the target compound with sufficient solubility
in water (this was discovered experimentally). This was
achieved by attaching compound 3-acid to the N-terminus of
Results and discussion
Synthesis and characterisation
For the purposes of this study, we targeted a series of com- the globally protected RGD peptide to give compound 4 in
pounds in which an RGD peptide unit was attached to 52% yield. Reduction of the azide to the primary amine using
different hydrophobic groups (Fig. 1) – we intended to demon- H2 with Pd/C proceeded to provide compound 5 in quantitative
strate the impact of the hydrophobic (self-assembling) part of yield. Finally, coupling compound 5 to compound 2-acid via
the molecule on self-assembly.
TBTU-mediated peptide coupling, followed by global
Fig. 1 Compounds investigated for self-assembly and integrin binding in this paper – C12-RGD, Py-RGD, C22-RGD and (C12)2Lys-RGD.
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2013

View Article Online
Organic & Biomolecular Chemistry
Paper
Scheme 2 Synthesis of (C12)2Lys-RGD.
343 nm, λem = 380–440 nm). On increasing the concentration
of Py-RGD the fluorescence emission spectrum decreased in
intensity, indicating that self-assembly of pyrene in the core of
the self-assembled nanostructures results in fluorescence
quenching. Plotting the monomer emission (at 382 nm)
against [Py-RGD] indicated that at ca. 120 μM, there was a dis-
continuity (see ESI† for data) – this reflects the CAC value, at
which assembly into nanostructures becomes favoured. There-
fore, the data collected using the inherent fluorescence of
Py-RGD was in general agreement with the Nile Red ‘external
dye’ approach.
Table 1 Critical Aggregation Concentration (CAC) values determined by Nile
Red encapsulation assay
Compound
CAC value/μM
C12-RGD
Py-RGD
C22-RGD
295
110
30
(C12)2Lys-RGD
6
deprotection gave target compound (C12)2Lys-RGD in an
acceptable yield of 11%.
On further increasing the extent of hydrophobic functiona-
lisation, the CAC changed by an order of magnitude, with self-
assembly becoming significantly more favoured. For com-
pound C22-RGD the CAC was determined as 30 μM using the
Nile Red assay – a ten-fold reduction of CAC compared with
C12-RGD, indicating the effect of extending the hydrophobic
chain length. In our previous work on polycationic self-assem-
blies for DNA binding, we have also noted that increasing the
hydrophobic chain length from C12 to C22 induced a change
in CAC in a similar way, although even more dramatically from
210 μM to 2 μM.5f We suggest that the magnitude of the effect
of hydrophobic functionalisation on the CAC of self-assembled
systems depends on the amount of charge on the head group
which has to be brought into close proximity on the surface of
the self-assembled nanostructure.
Self-assembly of RGD derivatives
The self-assembly of the four target compounds was initially
investigated via a Nile Red assay – in this experiment, Nile Red
only becomes solubilised into water (pH 7.4) at concentrations
above the critical aggregation concentration (CAC).16 Monitor-
ing the fluorescence emission of Nile Red (λex = 550 nm, λem
=
650 nm), and plotting the intensity against log[RGD-derivative]
(see ESI†) allows the CACs to be determined (Table 1). Under
these pH conditions, the compounds investigated should have
one net negative charge on the RGD unit (the arginine side
chain is cationic, while the aspartic acid is anionic on both its
side chain and its terminus).
Interestingly, Py-RGD had
a significantly lower CAC
Finally, twin-tailed (C12)2Lys-RGD exhibited highly effective
self-assembly, with a very low CAC of 6 μM, once again indicat-
ing that increasing the degree of hydrophobic functionalisa-
tion drives increased propensity for aggregation in aqueous
solution. The thermodynamics of self-assembly were explored
further using multiscale modelling methods (see below).
(110 μM) than C12-RGD (295 μM). We suggest that π–π inter-
actions and packing effects between pyrene groups may
reinforce the self-assembly of these hydrophobic units. In
order to confirm this suggestion, we carried out variable con-
centration experiments on Py-RGD in order to probe the
inherent fluorescent emission from the pyrene units (λex
=
This journal is © The Royal Society of Chemistry 2013
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
Morphology of self-assembly
that these objects were hollow, but also in some places cylin-
der bundling was possible. In order to understand this mor-
The pioneering work of Israelachvili states that the hydro- phology more fully, multiscale modelling was required (see
philic/lipophilic balance of surfactant-like compounds con- below for comments). Clearly, however, extension of the hydro-
trols the morphology of their self-assembly,17 and we therefore phobic chain has changed the mode of self-assembly in agree-
probed the morphology of our self-assembled nanostructures ment with the work of Israelachvili,17 who noted that if the
using both experimental and theoretical methods.
hydrophobic balance of the molecule increases there is no
Transmission electron microscopy was used to visualise the longer sufficient ‘space’ within a spherical micelle to accom-
nanoscale assemblies (Fig. 2). In each case, solutions of the modate the hydrophobic unit.
self-assemblies were dried on the TEM grid, negatively stained
Interestingly, C22-RGD formed gels in d6-DMSO – it is well-
with uranyl acetate and then visualised. TEM is a useful com- known that systems which can self-assemble into fibrillar
parative technique for families of related molecular structures nanostructures can form self-supporting networks which
(as in this case). Compound C12-RGD exhibited clusters of underpin supramolecular gels if the solvent is well-chosen.18
micellar objects when observed by TEM. These micelles were In this case, although a gel formed in d6-DMSO, the system
of ca. 5 nm diameter, which is in accordance with the head-to- did not gel in aqueous solution, however, there was increased
tail molecular length of this compound, which is ca. 2.5 nm. viscosity.
Interestingly, if C12-RGD was dried from a more dilute solu-
Twin tailed compound, (C12)2Lys-RGD, was also observed
tion ( just above the CAC), these micelles were not observed, to form a rod-like self-assembly (Fig. 2D). In this case,
and this may indicate that the self-assembled nanostructures however, isolated rods could be observed (Fig. 3). It should be
have relatively low stability in this case. Similarly, compound noted that the SEM image (Fig. 2D) is dried from a signifi-
Py-G2 also exhibited clustered micellar assemblies. However, cantly more concentrated solution than the TEM image in
unlike for C12-RGD, if Py-RGD was dried from just above its Fig. 3A, and as such, the apparent network densities are some-
CAC value, micelles were still observed, which is suggestive of what different in the two images. The rods were monodisperse
these self-assembled structures having greater stability.
with diameters of ca. 11.5 nm. Given that the extended head-
Compound C22-RGD exhibited completely different to-tail molecular length is ca. 6 nm, we proposed that the
a
morphology with cylindrical assemblies being proposed as the structures observed by TEM are probably simple cylindrical
dominant assembly mode (see computational modelling micelles in this case.
below for further information). However, the dimensions of
In order to gain further insight into self-assembly we went
the observed objects were ca. 20 nm – in contrast to the mole- on to perform computational modelling. Modelling has pre-
cular-scale diameter of ca. 4.1 nm. There was some evidence viously been applied by other groups to understand the self-
Fig. 2 TEM images of samples dried from solutions with concentrations of 1 mM. A: C12-RGD, scale bar = 50 nm; B: Py-RGD, scale bar = 100 nm; C: C22-RGD,
scale bar = 500 nm; D: (C12)2Lys-RGD, scale bar = 200 nm. All samples negatively stained with uranyl acetate, except image C.
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2013

View Article Online
Organic & Biomolecular Chemistry
Paper
Fig. 3 TEM images of compound (C12)2Lys-RGD dried from a concentration
of 100 mM, with scale bars of A: 500 nm and B: 50 nm. Image B clearly shows
the dimensions of an individual self-assembled cylindrical micelle.
assembly of RGD peptide-amphiphiles.19 Full details of com-
putational methods can be found in the ESI,† but the methods
can be summarised as follows. Initially we optimized the struc-
tures of the individual molecules using AMBER 1120 and the
GAFF force field21 in a TI3P water box22 at proper pH and ionic
strength using molecular dynamics methods. The SHAKE
Fig. 4 Mesoscale modeling of the amphiphilic RGD peptides derivatives syn-
thesized in this work showing self-assembly into micellar objects of different
morphologies: spherical micelles from C12-RGD (top panel, left) and Py-RGD
(top panel, right), cylindrical micelle from (C12)2Lys-RGD (middle panels, front
algorithm was employed,23 with the Berendsen coupling algori- and lateral view), and rod-like vesicles from C22-RGD (bottom panels, front and
thm providing temperature control24 and the particle mesh
lateral view). In all pictures, the dark grey sticks represent the RGD hydrophilic
units. Different coloured sticks are adopted to represent the various hydrophobic
Ewald method to treat long range electrostatics.25 Well estab-
regions (see Experimental and Computational methods for more details). Light
lished procedures were then used to calculate the interaction
blue spheres are used to portray some representative water molecules.
energies within the solvated system.26 We subsequently para-
meterised the molecules in order to apply mesoscale dissipa-
tive particle dynamics (DPD) methods27 and hence probe their
aggregate into supramolecular structures, with different mor-
phologies dependent on the precise molecular architecture.
Specifically: (i) C12-RGD and Py-RGD assemble into spherical
micelles, with average diameter (Dm) and aggregation number
(N) of 4.6 nm and 52 for C12-RGD and 4.1 nm and 27 for Py-
RGD, respectively, (ii) the double-tailed compound (C12)2Lys-
RGD self-assembles into an extended cylindrical micelle with
Dm = 9.8 nm and N = 3200, and (iii) the long hydrophobic
chain-bearing amphiphile C22-RGD generates rod-like vesicles
characterized by a micellar diameter value (Dm) of 23.1 nm
and an aggregation number (N) of 3050. As such, there
appears to be an excellent degree of agreement between the
theoretical (Fig. 4) and experimental (Fig. 2 and 3) obser-
vations. In particular, the ability of modelling to shed light on
the rod-like vesicle morphology formed by C22-RGD is particu-
larly noteworthy. The first two columns of Table 2 summarize
these geometrical findings in numerical form.
self-assembly behaviour on the nanometre scale. The DPD
model captures the essentials of mass and momentum conser-
vation which are responsible for the hydrodynamic behaviour
of a fluid at large scales and treats the molecules as a series of
beads connected together by Hooekean springs. Further
details of the number and choice of beads and methods
employed to determine these5d,28 can be found in the ESI.†
Electrostatic interactions between beads were analysed follow-
ing the approach outlined by Groot,29 and solvent molecules
were also simulated. The characteristic time scale for the
systems was then defined matching the experimental and
simulated diffusivity for water molecules.30 As such the model-
ling approach applied is multiscale, ideal for simulating self-
assembled nanostructures in solution, and includes aspects of
both molecular dynamics and mesoscale modelling.
We began our DPD-based simulation by monitoring the
molecular self-assembly processes to gain an insight into the
aggregates formed. Fig. 4 summarises the results of the model-
ling and shows that all hydrophobically-modified RGDs
Considering the origins of this effect in more detail, two
opposing forces control self-assembly: hydrocarbon–water
interactions that favour aggregation, and head group
This journal is © The Royal Society of Chemistry 2013
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
Table 2 Values of micellar diameter Dm (nm), core radius Rc (nm), aggregation
number N, and packing parameter P for the different modified RGD compounds
as obtained from DPD simulations
Table 3 Predicted free energy of micellisation ΔGm (kJ mol−1) and critical
micelle concentration CMC (μM) for the different modified RGD compounds
Compound
ΔGm
CMC
Compound
Dm
4.6 0.2
4.1 0.1
23.1 0.5
9.8 0.4
Rc
N
P
C12-RGD
Py-RGD
C22-RGD
(C12)2Lys-RGD
−10.37
−11.00
−12.73
−13.70
157
92.2
21.3
9.4
C12-RGD
Py-RGD
C22-RGD
(C12)2Lys-RGD
1.4
1.2
—
52
27
3050
3200
0.32
0.30
0.80
0.43
2.2
The critical micelle concentration (CMC) is one of the most
interactions that work in the opposite sense. These two contri- commonly studied properties of a self-assembled system
butions can be considered as an attractive interfacial tension because it is a direct measure of the thermodynamic stability
term due to hydrocarbon tails and a repulsion term depending of the micelles in solution. Basic thermodynamic relationships
on the nature of the hydrophilic group. This basic idea was allow CMC to be directly obtained from ΔGm; in particular, for
originally quantified by Israelachvili, Mitchell and Ninham,17 ionic micelles in a solution of high electrolyte content as in
resulting in the concept that aggregation of surfactants is con- the present study, we have:
trolled by molecular geometry.
ΔGm ꢀ RTlnðCMCÞ
ð1Þ
The packing parameter, P,17 is a very useful parameter
which allows the prediction of aggregate shape and size; as is
well-known, P values < 0.33 lead to spherical or ellipsoidal
micelles, when 0.33 < P < 0.5 large cylindrical or rod-shape
micelles are predicted, while flexible bilayer structures and
vesicles are predicted when 0.5 < P < 1 and, lastly, reversed or
inverted micelles appear for P > 1. Simple modelling consider-
ations allowed us to calculate the packing parameter P for all
of the modified RGD compounds in hydrated conditions, as
shown in the last column of Table 2. Pleasingly, the calculated
P values are in agreement with the more detailed multiscale
modeling studies (and experimental observations) about the
precise way in which the hydrophobic tails control the ultimate
self-assembled morphology of these RGD derivatives.
The second column in Table 3 shows the calculated CMC
data for each system simulated in this work. Typically, micellar
aggregates have CMCs of the order of 10−3–10−5 M, while lower
CMCs are found for amphiphiles that form membranes or
cylindrical aggregates. This is, at least in part, a consequence
of the fact that amphiphiles with low CMCs tend to have rela-
tively large hydrophobic segments, and this normally results
in an assembly shape with a lower curvature. Although a word
of caution is due about the fact that the calculated values of
ΔGm and CMC are obtained using simplified (but validated)
theoretical approaches, importantly the trends exhibited by
these parameters are pleasingly in-line with the CAC experi-
mental data (compare values in Tables 1 and 3).
From an energetic standpoint, the change in Gibbs free
energy of transfer of a single amphiphilic molecule from the
monomeric state to a micelle of aggregation number N, com-
monly called the free energy of micellisation ΔGm, can be
modelled as consisting of a hydrophobic part, ΔGm,h, and an
electrostatic part, ΔGm,e, so that ΔGm = ΔGm,h + ΔGm,e. The
value ΔGm,h is essentially the free energy associated with trans-
ferring hydrocarbon chains out of water and into the oil-like
interior of the micelle (a favourable contribution), plus the
surface free energy attributed to solvent-hydrocarbon contacts
in the micelle (an unfavourable contribution). The second
term, ΔGm,e originates from unfavourable contributions
associated with head group interactions, including electro-
static as well as head group conformation effects. We calcu-
lated the values of ΔGm for the four RGD-derivatives as shown
in the first column of Table 3. As expected, ΔGm at room temp-
erature and physiological ionic strength, has large, negative
values, indicating that micellisation is a spontaneous and
highly favourable process for all compounds. Interestingly,
however, the predicted values of ΔGm decrease on going from
C12-RGD to C22-RGD and, since the headgroup architecture
is the same in all amphiphiles, the main differential contri-
bution to ΔGm stems from the ΔGm,h term which, in turn,
must reflect differences in the size and structure of the hydro-
phobic component.
The most important aspect of our simulation approach is
the fact that the geometrical/energetical values of the different
micellar entities predicted by our computational recipe are in
excellent agreement with the corresponding values stemming
from experimental evidence (TEM, CAC); confirming the
relevance of multiscale molecular modeling and simulation in
the study of self-assembly of complex molecular entities. Last
but certainly not least, modelling strongly suggests that, for
C22-RGD, the fibres observed by TEM actually consist of
hollow lamellar-type assemblies rather than bundles of indi-
vidual fibrils. In this sense, the modelling provides additional
insight into the mode of self-assembly which was difficult to
obtain using the experimental methods employed, once again
demonstrating the synergy between these approaches.
Integrin binding studies
We then went on to monitor the binding of these self-assem-
blies to integrin at pH 7.4, using a fluorescence polarisation
(FP) assay, as previously reported by others and ourselves.15,31
In this assay, integrin αvβ3, dispersed in Triton X-100, is bound
to a fluorescent probe molecule (a fluorescein-modified cyclic
RGD peptide synthesised in-house).15 The compound of inter-
est is then introduced, and if it binds to the integrin, the fluore-
scent probe is displaced, which can be detected by a decrease
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2013

View Article Online
Organic & Biomolecular Chemistry
Paper
like a very weakly binding monovalent RGD peptide ((C12)2Lys-
RGD). It is possible that in these cylindrical-type aggregates,
the RGD peptide ligands are less-accessible on the surfaces of
the cylinders. Interestingly, Lee and co-workers previously
reported that nanostructure morphology could affect the
binding of multivalent mannose units to concanavalin A, with
small spherical micellar aggregates being shown to have more
effective binding to the target than cylindrical aggregates.4d
Stupp and co-workers have also previously reported that the
mobility and accessibility of peptide ligands on self-assembled
multivalent surfaces is important in controlling their ability to
bind to biological partners.14e These previous observations
would therefore appear be in agreement with our observations
as spherical micelles might, as a consequence of their geome-
try, be expected to provide more space on their surfaces for
each individual ligand to bind to its biological partner (accord-
ing to Israelachivili’s geometric considerations). Alternatively,
it is possible that the higher aggregation numbers of the
cylindrical aggregates (see Table 2) lowers the on-off rate of
binding at their surfaces.
Fig. 5 Data from integrin binding assay showing effect of RGD-derivatives on
the fluorescence polarisation signal of a fluorescent probe bound to intergin
αvβ3 in Triton X-100. Orange = C12-RGD, Blue = Py-RGD, Red = C22-RGD,
Purple = (C12)2Lys-RGD. All samples were incubated at 29 °C for 5 min.
in the FP signal associated with the enhanced mobility of the
probe. This assay is a useful way of comparing related ligands,
and gaining a degree of quantitative insight into their inter-
actions with integrin.
However, it should also be noted that in both cases, the vis-
cosity of the solution increased on adding these cylindrical
assemblies, which have a tendency towards the formation of
entangled gel-phase network. Fluorescence polarisation assays
are highly sensitive to viscosity increases, and so this may also
explain why the cylindrical nanoscale morphologies are not as
effective as the spherical micelles in this integrin binding
assay. Fig. 6A is a TEM image of the Triton X-100-integrin
As previously reported,15 C12-RGD showed a good affinity
for integrin αvβ3 in this assay, and reduced the FP signal to
below 50% of its initial level at a concentration of 200 μM
(Fig. 5). We therefore define the EC50 as 200 μM for this
ligand. This is significantly more effective than a non-self-
assembling RGD peptide, as we reported previously which was
unable to displace 50% of the probe from the integrin binding
site. In order to ensure that any effect on the process was not
simply due to the surfactant interfering with the Triton X-100
present in the assay through a detergency effect, we used
sodium dodecyl sulphate (SDS) as a negative control. This
compound led to no change in the FP signal demonstrating
that the presence of the RGD peptide was essential, indicative
of specific interactions between the self-assembling peptide
and the integrin protein binding site.
We then tested Py-RGD in the same assay. This compound
exhibited an EC50 for integrin binding of 110 μM (Fig. 5). Once
again, this is evidence for a self-assembled multivalency effect
on binding with the micellar assembly being able to bind to
integrin proteins in a more effective way. Intriguingly, the EC50
value for Py-RGD is very similar to its CAC, and we suggest that
only once the peptide assembles can really effective multi-
valency, and hence effective protein binding be observed. We
note, in agreement with this hypothesis, that Py-RGD is a
more effective integrin binder than C12-RGD and that the
formed compound also has a lower CAC value, and hence will
more easily be able to express self-assembled multivalent
binding.
We then went on to test the ability of C22-RGD and
(C12)2Lys-RGD to interact with integrin αvβ3 in the same assay.
These compounds were unable to effectively bind to the
protein and exhibited either no binding (C22-RGD) or behaved
Fig. 6 TEM images of A: Triton X-100/Integrin, scale bar = 500 nm; B: Self-
assembled fibres of C22-RGD in the presence of Triton X-100/Integrin, scale bar
= 5 μm. Image 6A reproduced from ref. 15.
This journal is © The Royal Society of Chemistry 2013
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
system used in the assay, and Fig. 6B shows the cylindrical morphologies in a biological setting and developing more
micelles formed by C22-RGD at the end of the integrin detailed structure–activity relationships for SAMul binding
binding assay. The Triton X-100/integrin assemblies some of arrays.
which can be seen tangled up within the fibrous network
formed by C22-RGD.
Nonetheless, we conclude that for effective self-assembled
multivalent solution-phase binding of integrin proteins in this
Acknowledgements
chemical assay, it is preferable to have spherical micellar DKS acknowledges EPSRC and University of York for funding
nanoscale assemblies, rather than their cylindrical equi- (through the DTA programme). PP and SP wish to acknowledge
valents. In this way, the nanoscale morphology programmed the financial support from ESTECO through the project DDOS.
in by self-assembly controls the outcome of the binding assay. Part of this work was carried out under the HPC-Europa 2
Of course, it should be noted that in biological systems, these project MONALISA (CINECA Supercomputing Center, Bologna,
binding effects may differ, as the integrin proteins will be dis- Italy), funded by the European Commission – DG Research in
played on cell surfaces in a very different manner to the surfac- the 7th Framework Program (Grant agreement no. 228398). All
tant-immobilised proteins within this assay. Clearly this is authors acknowledge the support of the COST program,
worthy of further study. However, this paper clearly illustrates TD0802 (Dendrimers in Biomedicine) for support of network-
the impact the effect which morphology can have on nanoscale ing activities.
binding events.
Notes and references
Conclusions
1 (a) M. Mammen, S. K. Choi and G. M. Whitesides, Angew.
We have reported the synthesis of four different self-assem-
bling RGD-peptide derivatives. Dependent on the choice of the
hydrophobic unit, the critical aggregation concentration can
be varied over several orders of magnitude, and the nanoscale
morphology into which the peptides assemble can be con-
trolled to be spherical, cylindrical or tubular. Multiscale mod-
elling provided direct insight into the self-assembly of these
ligands and predicted the precise degree of morphological
control, as well as predicting their critical aggregation concen-
trations. The multiscale modelling allowed us to dissect the
thermodynamic terms responsible for self-assembly, and
locate the origin of the morphological control firmly in the
nature of the interactions between the hydrophobic units
within the self-assembled nanostructures.
Chem., Int. Ed., 1998, 37, 2755–2794; (b) S. E. Stiriba,
H. Frey and R. Haag, Angew. Chem., Int. Ed., 2002, 41, 1329–
1334; (c) A. Mulder, J. Huskens and D. N. Reinhoudt, Org.
Biomol. Chem., 2004, 2, 3409–3424; (d) J. D. Badjic,
A. Nelson, S. J. Cantrill, W. B. Turnbull and J. F. Stoddart,
Acc. Chem. Res., 2005, 38, 723–732; (e) V. Martos,
P. Castreño, J. Valero and J. de Mendoza, Curr. Opin. Chem.
Biol., 2008, 12, 698–706; (f) G. M. Pavan, A. Danani, S. Pricl
and D. K. Smith, J. Am. Chem. Soc., 2009, 131, 9686–9694;
(g) S. P. Jones, G. M. Pavan, A. Danani, S. Pricl and
D. K. Smith, Chem.–Eur. J., 2010, 16, 4519–4532.
2 (a) A. Barnard and D. K. Smith, Angew. Chem., Int. Ed.,
2012, 51, 6572–6581; (b) K. Petkau-Milroy and L. Brunsveld,
Org. Biomol. Chem., 2013, 11, 219–232.
On studying the ability of these self-assembled nano-
structures to bind in a multivalent manner to integrin pro-
teins, it was found that the spherical micellar systems were far
more effective in our chemical solution-phase integrin binding
assays, than their cylindrical analogues. Furthermore, they
were much more effective than a non-self-assembling ana-
logue. Interestingly, the integrin binding behaviour of the
spherical micelles only appeared to become effective once the
compound was above its critical aggregation concentration –
providing further evidence for the conclusion that that self-
assembly plays a vital role in enabling high-affinity integrin
binding – self-assembled multivalency (SAMul).2 The cylindri-
cal, rod-like assemblies led to viscosification, and appeared
unable to displace the fluorescent probe from the integrin
biding site. As such, this paper indicates the importance of
optimising nanoscale morphology and dynamics using a com-
bined experimental and theoretical methodology when gener-
ating SAMul ligand arrays, and clearly demonstrates the
versatility and controllability of this approach. Future work
will focus on investigating the performance of different
3 (a) G. M. Whitesides, J. P. Mathias and C. T. Seto, Science,
1991, 254, 1312–1319; (b) J.-M. Lehn, Proc. Natl. Acad.
Sci. U. S. A., 2002, 99, 4763–4768; (c) J.-M. Lehn, Science,
2002, 295, 2400–2403; (d) G. M. Whitesides, Small, 2005, 1,
172–179; (e) D. A. Uhlenheuer, K. Petkau and L. Brunsveld,
Chem. Soc. Rev., 2010, 39, 2817–2826.
4 (a) J. E. Kingerywood, K. W. Williams, G. B. Sigal and
G. M. Whitesides, J. Am. Chem. Soc., 1992, 114, 7303–7305;
(b) G. Thoma, A. G. Katopodis, N. Voelcker, R. O. Duthaler
and M. B. Streiff, Angew. Chem., Int. Ed., 2002, 41, 3195–
3198; (c) A. Nelson, J. M. Belitsky, S. Vidal, C. S. Joiner,
L. G. Baum and J. F. Stoddart, J. Am. Chem. Soc., 2004, 126,
11914–11922; (d) B. S. Kim, D. J. Hong, J. Bae and M. Lee,
J. Am. Chem. Soc., 2005, 127, 16333–16337; (e) J.-H. Ryu,
E. Lee, Y.-B. Lim and M. Lee, J. Am. Chem. Soc., 2007, 129,
4808–4814; (f) M. K. Muller and L. Brunsveld, Angew.
Chem., Int. Ed., 2009, 48, 2921–2924; (g) J. Voskuhl,
M. C. A. Stuart and B. J. Ravoo, Chem.–Eur. J., 2010, 16,
2790–2796; (h) J. Geng, F. Biedermann, J. M. Zayed, F. Tian
and O. A. Scherman, Macromolecules, 2011, 44, 4276–4281.
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2013

View Article Online
Organic & Biomolecular Chemistry
Paper
5 (a) D. Joester, M. Losson, R. Pugin, H. Heinzelmann, 12 (a) L. J. Kornberg, H. S. Earp, C. E. Turner, C. Prockop and
E. Walter, H. P. Merkle and F. Diederich, Angew. Chem., Int.
Ed., 2003, 42, 1486–1490; (b) M. Guillot, S. Eisler, K. Weller,
H. P. Merkle, J. L. Gallani and F. Diederich, Org. Biomol.
Chem., 2006, 4, 766–769; (c) S. P. Jones, N. P. Gabrielson,
D. W. Pack and D. K. Smith, Chem. Commun., 2008, 4700–
4702; (d) P. Posocco, S. Pricl, S. Jones, A. Barnard and
D. K. Smith, Chem. Sci., 2010, 1, 393–401; (e) S. P. Jones,
N. P. Gabrielson, C. H. Wong, H. F. Chow, D. W. Pack,
P. Posocco, M. Fermeglia, S. Pricl and D. K. Smith, Mol.
L. Juliano, Proc. Natl. Acad. Sci. U. S. A., 1991, 88, 8392–
8396; (b) S. Miyamoto, S. K. Akiyama and K. M. Yamada,
Science, 1995, 267, 883–885; (c) G. Maheshwari, G. Brown,
D. A. Lauffenburger, A. Wells and L. J. Griffith, J. Cell Sci.,
2000, 113, 1677–1686; (d) D. C. Worth and M. Parsons,
Int. J. Biochem. Cell Biol., 2008, 40, 2397–2409; (e) L. Auzzas,
F. Zanardi, L. Battistini, P. Burreddu, P. Carta, G. Rassu,
C. Curti and G. Casiraghi, Curr. Med. Chem., 2010, 17,
1255–1299.
Pharm., 2011, 8, 416–429; (f) A. Barnard, P. Posocco, 13 (a) B. Hu, D. Finsinger, K. Peter, Z. Guttenberg,
S. Pricl, M. Calderon, R. Haag, M. E. Hwang,
V. W. T. Shum, D. W. Pack and D. K. Smith, J. Am. Chem.
Soc., 2011, 133, 20288–20300; (g) S. K. M. Nalluri,
J. Voskuhl, J. B. Bultema, E. J. Boekema and B. J. Ravoo,
Angew. Chem., Int. Ed., 2011, 50, 9747–9751;
(h) N. P. Gabrielson and J. J. Cheng, Biomaterials, 2010, 31,
9117–9127.
M. Barmann, H. Kessler, A. Escherich, L. Moroder,
J. Bohm, W. Baumeister, S. Sui and E. Sackmann, Biochem-
istry, 2000, 39, 12284–12294; (b) V. Marchi-Artzner, B. Lorz,
U. Hellerer, M. Kantlehner, H. Kessler and E. Sackmann,
Chem.–Eur. J., 2001, 7, 1095–1101; (c) S. Cressman,
I. Dobson, J. B. Lee, Y. Y. C. Tam and P. R. Cullis, Bioconju-
gate Chem., 2009, 20, 1404–1411; (d) V. Marchi-Artzner,
B. Lorz, C. Gosse, L. Jullien, R. Merkel, H. Kessler and
E. Sackmann, Langmuir, 2003, 19, 835–841; (e) M. Colin,
R. P. Harbottle, A. Knight, M. Kornprobst, R. G. Cooper,
A. D. Miller, G. Trugnan, J. Capeau, G. Coutelle and
M. C. Brahimi-Horn, Gene Ther., 1998, 5, 1488–1498;
(f) M. F. M. Mustapha, S. M. Grosse, L. Kudsiova, M. Elbs,
E. A. Raiber, J. B. Wong, A. P. R. Brain, H. E. J. Armer,
A. Warley, M. Keppler, T. Ng, M. J. Lawrence, S. L. Hart,
H. C. Hailes and A. B. Tabor, Bioconjugate Chem., 2009, 20,
518–532; (g) J. A. Zupancich, F. S. Bates and
M. A. Hillmeyer, Biomacromolecules, 2009, 10, 1554–1563;
(h) Z. Zhang, Y. Lai, L. Yu and J. Ding, Biomaterials, 2010,
31, 7873–7882; (i) P. M. Valencia, M. H. Hanewich-Hollatz,
W. Gao, F. Karim, R. Langer, R. Karnik and
O. C. Farokhzad, Biomaterials, 2011, 32, 6226–6233;
( j) Y. Wang, J. Wu, G. Kang, M. Zhao, L. Gui, N. Li, L. Peng,
X. Zhang, L. Li and S. Peng, J. Mater. Chem., 2012, 22,
4652–4659; (k) E. Kluza, D. W. J. van der Schaft,
P. A. I. Hautvast, W. J. M. Mulder, K. H. Mayo,
A. W. Griffioen, G. J. Strijkers and K. Nicolay, Nano Lett.,
2010, 10, 52–58; (l) E. Kluza, I. Jacobs, S. J. C. G. Hectos,
K. H. Mayo, A. W. Griffioen, G. J. Strijkers and K. Nicolay,
J. Controlled Release, 2012, 158, 207–214.
6 A. C. Rodrigo, A. Barnard, J. Cooper and D. K. Smith,
Angew. Chem., Int. Ed., 2011, 50, 4675–4679.
7 For example reviews see: (a) G. Colombo, P. Soto and
E. Gazit, Trends Biotechnol., 2007, 25, 211–218;
(b) Y.-C. Tseng and L. Huang, J. Biomed. Nanotechnol., 2009,
5, 351–363; (c) A. M. Carmona-Ribeiro, Int. J. Nanomed.,
2010, 5, 249–259; (d) R. Chhabra, J. Sharma, Y. Liu,
S. Rinker and H. Yan, Adv. Drug Delivery Rev., 2010, 62,
617–625; (e) J. Shi, Z. Xiao, N. Kamaly and O. C. Farokhzad,
Acc. Chem. Res., 2011, 44, 1123–1134.
8 (a) R. O. Hynes, Cell, 1992, 69, 11–25; (b) O. Cleaver and
D. A. Melton, Nat. Med., 2003, 9, 661–668; (c) E. F. Plow,
T. A. Haas, L. Zhang, J. Loftus and J. W. Smith, J. Biol.
Chem., 2000, 275, 21785–21788; (d) R. O. Hynes, Nat. Med.,
2002, 8, 918–921.
9 (a) M. D. Pierschbacher and E. Ruoslahti, Nature, 1984,
309, 30–33; (b) J.-P. Xiong, T. Stehle, R. Zhang,
A. Joachimiak, M. Frech, S. L. Goodman and M. A. Arnaout,
Science, 2002, 296, 151–155; (c) M. A. Dechantsreiter,
E. Planker, B. Matha, E. Lohof, G. Holzemann, A. Jonczyk,
S. L. Goodman and H. Kessler, J. Med. Chem., 1999, 42,
3033–3040; (d) G. Tabatabai, M. Weller, B. Nabors,
M. Picard, D. Reardon, T. Mikkelsen, C. Ruegg and
R. Stupp, Targ. Oncol., 2010, 5, 175–181.
14 (a) J. A. Zupancich, F. S. Bates and M. A. Hillmyer, Bioma-
cromolecules, 2009, 10, 1554–1563; (b) Y.-B. Lim, E. Lee and
M. Lee, Angew. Chem., Int. Ed., 2007, 46, 3475–3478;
(c) Y.-B. Lim, O.-J. Kwon, E. Lee, P.-H. Kim, C.-O. Yun and
M. Lee, Org. Biomol. Chem., 2008, 6, 1944–1948;
(d) M. O. Guler, L. Hsu, S. Soukasene, D. A. Harrington,
J. F. Hulvat and S. I. Stupp, Biomacromolecules, 2006, 7,
1855–1863; (e) H. Storrie, M. O. Guler, S. N. Abu-Amara,
T. Volberg, M. Rao, B. Geiger and S. I. Stupp, Biomaterials,
2007, 28, 4608–4618; (f) Z. Huang, T. D. Sargeant,
J. F. Hulvat, A. Mata, P. Bringas, C.-Y. Koh, S. I. Stupp and
M. L. Snead, J. Bone Miner. Res., 2008, 23, 1995–2006;
(g) M. Zhou, A. M. Smith, A. K. Das, N. W. Hodgson,
R. F. Collins, R. V. Ulijn and J. E. Gough, Biomaterials,
2009, 30, 2523–2530; (h) L. Zhang, F. Rakotondradany,
10 L. Perlin, S. MacNeil and S. Rimmer, Soft Matter, 2008, 4,
2331–2349.
11 (a) P. C. Brooks, R. A. Clark and D. A. Cheresh, Science,
1994, 264, 569–571; (b) P. C. Brooks, A. M. P. Montgomery,
M. Rosenfeld, R. A. Reisfeld, T. Hu, G. Klier and
D. A. Cheresh, Cell, 1994, 79, 1157–1164; (c) C. S. Chen,
M. Mrksich, S. Huang, G. M. Whitesides and
D. E. Ingber, Science, 1997, 276, 1425–1428; (d) W. Arap,
R. Pasqualini and E. Ruoslahti, Science, 1998, 279, 377–
380; (e) K. E. Gottschalk and H. Kessler, Angew. Chem.,
Int. Ed., 2002, 41, 3767–3774; (f) K. N. Sugahara,
T. Teesalu, P. P. Karmali, V. R. Kotamraju, L. Agemy,
D. R. Greenwald and E. Ruoslahti, Science, 2010, 328,
1031–1035.
This journal is © The Royal Society of Chemistry 2013
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
A. J. Myles, H. Fenniri and T. J. Webster, Biomaterials, 2009, 22 W. L. Jorgensen, J. Chandrasekhar, J. D. Madura,
30, 1309–1320; (i) R. Orbach, L. Adler-Abramovich,
S. Zigerson, I. Mironi-Harpaz, D. Seliktar and E. Gazit, Bio-
R. W. Impey and M. L. Klein, J. Chem. Phys., 1983, 79, 926–
935.
macromolecules, 2009, 10, 2646–2651; ( j) K. M. Galler, 23 J.-P. Ryckaert, G. Ciccotti and H. J. C. Berendsen, J. Comput.
L. Aulisa, K. R. Regan, R. N. D’Souza and J. D. Hartgerink, Phys., 1977, 23, 327–341.
J. Am. Chem. Soc., 2010, 132, 3217–3223; (k) G. Cheng, 24 H. J. C. Berendsen, J. P. M. Postma, W. F. Van Gunsteren,
V. Castelletto, R. R. Jones, C. J. Connon and
A. DiNola and J. R. Haak, J. Chem. Phys., 1984, 81, 3684–
I. W. Hamley, Soft Matter, 2011, 7, 1326–1333;
3690.
(l) J. Z. Gasiorowski and J. H. Collier, Biomacromolecules, 25 A. Toukmaji, C. Sagui, J. Board and T. Darden, J. Chem.
2011, 12, 3549–3558. Phys., 2000, 113, 10913–10927.
15 D. J. Welsh and D. K. Smith, Org. Biomol. Chem., 2011, 9, 26 (a) M. Fermeglia, M. Ferrone and S. Pricl, Fluid Phase
4795–4801.
Equilib., 2003, 212, 315–329; (b) R. Toth, A. Coslanich,
M. Ferrone, M. Fermeglia, S. Pricl, S. Miertus and
E. Chiellini, Polymer, 2004, 45, 8075–8083; (c) M. Fermeglia,
M. Ferrone and S. Pricl, Mol. Simul., 2004, 30, 289–300.
16 M. C. A. Stuart, J. C. van de Pas and J. B. F. N. Engberts,
J. Phys. Org. Chem., 2005, 18, 929–934.
17 (a) J. N. Israelachvili, D. J. Mitchell and B. W. Ninham,
J. Chem. Soc., Faraday Trans. 2, 1976, 72, 1525–1568; 27 (a) P. J. Hoogerbrugge and J. M. V. A. Koelman, Europhys.
(b) J. N. Israelachvili, D. J. Mitchell and B. W. Ninham,
Biochim. Biophys. Acta, 1977, 470, 185–201.
Lett., 1992, 19, 155–160; (b) J. M. V. A. Koelman and
P. J. Hoogerbrugge, Europhys. Lett., 1993, 21, 363–368;
(c) P. Español and P. Warren, Europhys. Lett., 1995, 30, 191;
(d) P. Español, Phys. Rev. E: Stat. Phys., Plasmas, Fluids,
Relat. Interdiscip. Top., 1995, 52, 1734; (e) C. Marsh,
G. Backx and M. H. Ernst, Phys. Rev. E: Stat. Phys., Plasmas,
Fluids, Relat. Interdiscip. Top., 1997, 56, 1976;
(f) P. B. Warren, Curr. Opin. Colloid Interface Sci., 1998, 3,
620; (g) R. D. Groot and P. B. Warren, J. Chem. Phys., 1997,
107, 4423.
18 (a) D. K. Smith, in Supramolecular Chemistry: From Mole-
cules to Nanomaterials, vol. 7, Soft Matter, ed. P. A. Gale and
J. W. Steed, John Wiley & Sons, Chichester, UK, 2012,
pp. 3355–3376; (b) Molecular Gels – Materials with Self-
Assembled Fibrillar Networks, ed. R. G. Weiss, P. Terech,
Springer, Dordrecht, 2006; (c) A. R. Hirst, B. Escuder,
J. F. Miravet and D. K. Smith, Angew. Chem., Int. Ed., 2008,
47, 8002–8801; (d) S. Banerjee, R. K. Das and U. Maitra,
J. Mater. Chem., 2009, 19, 6649–6687.
19 (a) S. E. Paramanov, H.-W. Jun and J. D. Hartgerink, J. Am.
Chem. Soc., 2006, 128, 7291–7298; (b) Y. S. Velichko,
S. I. Stupp and M. Olvera de la Cruz, J. Phys. Chem. B, 2008,
112, 2326–2334.
20 D. A. Case, T. A. Darden, T. E. Cheatham III,
C. L. Simmerling, J. Wang, R. E. Duke, R. Luo,
R. C. Walker, W. Zhang, K. M. Merz, B. Roberts, B. Wang,
S. Hayik, A. Roitberg, G. Seabra, I. Kolossváry, K. F. Wong,
F. Paesani, J. Vanicek, X. Wu, S. R. Brozell, T. Steinbrecher,
H. Gohlke, Q. Cai, X. Ye, J. Wang, M.-J. Hsieh, G. Cui,
D. R. Roe, D. H. Mathews, M. G. Seetin, C. Sagui, V. Babin,
T. Luchko, S. Gusarov, A. Kovalenko and P. A. Kollman,
AMBER 11, University of California, San Francisco, CA,
USA, 2010.
28 (a) R. Toth, F. Santese, S. P. Pereira, D. R. Nieto, S. Pricl,
M. Fermeglia and P. Posocco, J. Mater. Chem., 2012, 22,
5398–5409; (b) P. Posocco, M. Fermeglia and S. Pricl,
J. Mater. Chem., 2010, 20, 7742–7753; (c) M. Fermeglia and
S. Pricl, Comput. Chem. Eng., 2009, 33, 1701–1710;
(d) G. Scocchi, P. Posocco, J.-W. Handgraaf,
J. G. E. M. Fraaije, M. Fermeglia and S. Pricl, Chem.–Eur. J.,
2009, 15, 7586–7592; (e) R. Toth, D.-J. Voorn,
J.-W. Handgraaf, J. G. E. M. Fraaije, M. Fermeglia, S. Pricl
and P. Posocco, Macromolecules, 2009, 42, 8260–8270;
(f) G. Scocchi, P. Posocco, M. Fermeglia and S. Pricl,
J. Phys. Chem. B, 2007, 111, 2143–2151; (g) G. Scocchi,
P. Posocco, A. Danani, S. Pricl and M. Fermeglia, Fluid
Phase Equilib., 2007, 261, 366–374.
29 R. D. Groot, J. Chem. Phys., 2003, 118, 11265–11277.
21 (a) J. Wang, R. M. Wolf, J. Caldwell, P. A. Kollman and 30 R. D. Groot and K. L. Rabone, Biophys. J., 2001, 81, 725–
D. A. Case, J. Comput. Chem., 2004, 25, 1157–1174; 736.
(b) J. Wang, P. A. Kollman and D. A. Case, J. Mol. Graphics. 31 W. Wang, Q. Wu, M. Pasuelo, J. S. McMurray and C. Li, Bio-
Modell., 2006, 25, 247–260.
conjugate Chem., 2005, 16, 729–734.
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2013