Multivalent DNA recognition by self-assembled clusters: deciphering structural effects by fragments screening and evaluation as siRNA vectors

Article abstract of DOI:10.1039/c5ob01404b

The identification of low-molecular-weight clusters that effectively complex oligonucleotides of therapeutic interest is of great importance for applications in gene delivery. We recently reported the use of self-assembly processes based on chemoselective ligation in order to generate biomolecular clusters for the multivalent recognition of DNA. Herein, we exploit the modularity of this methodology to perform a one-pot fragments screening of scaffolds and binding groups. Structural parameters affecting DNA binding were observed and hits have been identified by fluorescence displacement and gel electrophoresis assays. Finally, we evaluated the potential of these systems for siRNA transfection. One biomolecular cluster was found to effectively complex and transport a 21-mer siRNA inside MCF7 human breast cancer cells, resulting in a significant knockdown of the target gene.

Full text of DOI:10.1039/c5ob01404b

Organic &
Biomolecular Chemistry
View Article Online
View Journal
PAPER
Multivalent DNA recognition by self-assembled
clusters: deciphering structural effects by
fragments screening and evaluation as siRNA
vectors†
Cite this: DOI: 10.1039/c5ob01404b
Eline Bartolami, Yannick Bessin, Nadir Bettache,* Magali Gary-Bobo, Marcel Garcia,
Pascal Dumy and Sébastien Ulrich*
The identification of low-molecular-weight clusters that effectively complex oligonucleotides of thera-
peutic interest is of great importance for applications in gene delivery. We recently reported the use of
self-assembly processes based on chemoselective ligation in order to generate biomolecular clusters for
the multivalent recognition of DNA. Herein, we exploit the modularity of this methodology to perform a
one-pot fragments screening of scaffolds and binding groups. Structural parameters affecting DNA
binding were observed and hits have been identified by fluorescence displacement and gel electropho-
resis assays. Finally, we evaluated the potential of these systems for siRNA transfection. One biomolecular
cluster was found to effectively complex and transport a 21-mer siRNA inside MCF7 human breast cancer
cells, resulting in a significant knockdown of the target gene.
Received 10th July 2015,
Accepted 31st July 2015
DOI: 10.1039/c5ob01404b
www.rsc.org/obc
Synthetic macromolecular vectors
Introduction
Challenges in gene therapy
Cationic polymers5 and dendrimers6 have been extensively
studied for the complexation and transport of oligonucleo-
The prospect of treating the genetic root of diseases is a fasci-
nating perspective of modern biotechnologies. Towards this
goal, synthetic oligonucleotides represent therapeutic agents
that can serve in antisense1 and silencing technologies.2
However, the effective and safe intracellular delivery of
synthetic oligonucleotides remains a tremendous challenge.
Oligonucleotides are negatively-charged biomolecules that are
relatively unstable in biological medium – they undergo enzy-
matic degradation by nucleases – and cannot pass cell mem-
branes due to electrostatic repulsion. Thus, the success of
oligonucleotide-based therapeutics depends greatly on the
development of effective vectors. The serious immunogenic
effects displayed by viral vectors3 call for an alternative
bottom-up chemical approach. The design of synthetic systems
that effectively complex and transport oligonucleotides is
therefore a topic of great importance.4
tides. In this case, the repetitions, within a single chain or
compound, of multiple binding units that promote complex
formation with oligonucleotides exert a multivalent7 binding.
However, the high molecular weight of these macromolecules
is an issue and toxic side-effects have been observed due to
their accumulation in biological tissues.8 A current approach
for lifting these limitations focus on the generation of low-
molecular-weight vectors that may be degradable and/or
stimuli-responsive.9,10
Multivalent DNA recognition by low-molecular-weight clusters
The multifunctionalization strategy (Scheme 1), whereby a
scaffold is functionalized with multiple copies of a binding
group through click-type reactions, has been recently exploited
for generating low-molecular-weight cationic clusters that
effectively complex oligonucleotides despite their limited
valency with respect to polymers.
Different examples of low-molecular-weight clusters based
on a variety of scaffolds (calixarene,11,12 fullerene,13 pillar[5]-
arene,14 cyclodextrines,15 and dendrimers16) and ligands
(ammonium, guanidinium) have thus been reported. These
systems represent promising alternatives to macromolecular
vectors, provided their structures are finely tuned in order
to optimize the multivalent binding to oligonucleotides.17
Institut des Biomolécules Max Mousseron (IBMM), UMR 5247, CNRS, Université
Montpellier, ENSCM, Ecole Nationale Supérieure de Chimie de Montpellier,
8 Rue de l’Ecole Normale, 34296 Montpellier cedex 5, France.
E-mail: nadir.bettache@univ-montp2.fr, Sebastien.Ulrich@enscm.fr
†Electronic supplementary information (ESI) available. See DOI: 10.1039/
c5ob01404b
This journal is © The Royal Society of Chemistry 2015
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
Scheme 1 Schematic representation of the principle of cluster for-
mation by multiple click ligations of a binding group (yellow triangle)
onto a functionalized scaffold.
Effective screening methods are therefore needed for facilitat-
ing the identification of potent oligonucleotide-binding
clusters.
Fig. 1 Structure of aldehyde and hydrazide building blocks that are
We recently reported the generation of tetravalent biomole-
cular clusters based on cyclic peptide scaffolds for the multi-
valent recognition of DNA.18 In comparison with the most
popular organic scaffolds, the use of biomolecules such as
peptides as scaffolds is of interest in terms of biocompatibility.
The selected bottom-up methodology rests upon the multi-
functionalization of the cyclic peptide scaffold through acyl-
hydrazone ligations. This type of ligation operates in mild
aqueous conditions and can tolerate the presence of a complex
biomolecular target such as DNA. Its chemoselectivity thus
enables tethering in one-pot unprotected binding groups onto
the scaffold. It is therefore a robust tool for programming the
assembly of multiple fragments into a bioactive cluster. It has
for instance been recently used by the Matile group for gene-
rating amphiphiles.19 The modularity of this fragment-assem-
bly process of cluster formation through acylhydrazone
ligations should enable the rapid screening of scaffolds and
binding groups which, in turn, would help develop a better
understanding of the molecular factors that determine the
DNA-binding ability of these clusters. We report herein the
results of a screening of different scaffolds and binding groups
that are assembled together in a combinatorial fashion. The
DNA-binding ability was directly analysed by a fluorescence-
displacement assay and cross-checked by gel electrophoresis.
Finally, the best hits were evaluated on cell culture as potential
siRNA vectors.
assembled through the acylhydrazone ligation.
Fig. 2 Synthetic route for the preparation of amino acid and peptide
hydrazides. (i) Fmoc-hydrazide, DIEA; (ii) (a) deprotection: piperidine, (b)
peptide coupling: Fmoc-amino acid-OH, HATU, DIEA, (c) deprotection:
piperidine; (iii) TFA/TIS/H2O.
tures and valency. The hydrazide partners are derivatives of
amino acids and tripeptides bearing a hydrazide group at
their C-terminus. They feature different types of side groups
(neutral, hydrophobic, anionic, cationic), different stereo-
chemistry, and different valency (compounds Hyd, Fig. 1).
The preparation of the linear peptide A7 was adapted from
the synthesis of the cyclic peptide A8,18 and involved solid-
phase peptide synthesis, acetylation of the N-terminal,
cleavage and deprotection, and oxidative cleavage of serine
side-chains to unmask the glyoxylic aldehydes.
The hydrazide partners were synthesized by a solid-phase
approach using a Fmoc-protected hydrazine resin that can be
readily prepared by reacting the 2-chlorotrityl chloride resin
with Fmoc-hydrazide (Fig. 2).20 The amino acid hydrazides
were prepared manually by peptide coupling, deprotection and
cleavage. The tripeptide hydrazides were assembled using
solid-phase peptide synthesis. After cleavage in mild acidic
conditions, the protected peptides were purified by reverse-
phase HPLC, before being fully deprotected to afford the tri-
peptide hydrazides.
Results and discussion
Design and synthesis of building blocks
The clusters are formed by tethering, through acylhydrazone
ligations, hydrazide binding groups onto functionalized
scaffolds featuring reactive aldehyde moieties (Scheme 1). The
selected functionalized scaffolds bear a varying number (func-
tionality = 1–4) of reactive aldehyde groups (compounds A,
Self-assembly of clusters by acylhydrazone ligation
Fig. 1). We selected different commercially-available com- The ligation reactions were carried out, as previously
pounds along with two synthetic – linear and cyclic – peptides. described, in aqueous buffer (100 mM AcONa, pH 5.0), with
Thus, these scaffolds feature different (bio)molecular struc- 1 mM of scaffolds A and 8 equivalents of hydrazide per
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Organic & Biomolecular Chemistry
Paper
scaffold. These conditions favour the formation of the bio-
conjugates A·Hyd of highest valency. To verify this and to
demonstrate that the ligation proceeds equally well with
respect to the different types of building blocks, we tested the
cluster formation of all hydrazides with scaffold A8 and of
AcHyd with all scaffolds A. HPLC and mass spectrometry ana-
lyses confirm in all cases the formation of the corresponding
clusters A·Hyd (see ESI†). These conditions are therefore suit-
able for the modular self-assembly of clusters from different
scaffolds and binding groups. We therefore envisaged carrying
out a one-pot screening for rapidly identifying the structural
determinants that favour multivalent DNA recognition.
Screening
Fig. 3 Screening of a combinatorial array of aldehyde scaffolds A (left)
and hydrazides binding groups (top) – assembled together by acylhydra-
zone ligations – by the ethidium bromide displacement assay with calf-
The ability of the different clusters to complex DNA was first
assessed by a fluorescence displacement assay with ethidium
bromide (EthBr) and calf thymus DNA (ctDNA). In this assay,
a fluorescent emission decrease upon addition of a compound
indicates its ability to complex DNA. This label-free assay is a
simple method for establishing DNA binding affinity.21 It is
also amenable to a high-throughput microplate format for
enabling screening approaches,22 and it has already been used
in this context for the determination of the DNA binding
sequence selectivity of a ligand.23 In this study, we selected
this assay for comparing the DNA-binding ability of clusters
formed from the combinatorial association of different frag-
ments (scaffolds and binding groups).
thymus DNA target. Relative fluorescence emission (λ
= 590 nm)
decrease: 0–25% ( ), 25–50% ( ), 50–75% ( ), 75–100% ( ). Measure-
ments performed 6 hours after the addition of the DNA target.
In each well of a 96-well plate, we deposited one aldehyde
scaffold A and one hydrazide building block (8 eq.). After over-
night mixing, acetate buffer (100 mM, pH 5.0) of low salinity
(9.4 mM NaCl) followed by ctDNA and EthBr were sub-
sequently added. DNA complexation by clusters results in a
fluorescence decrease that occurs within hours. The results
show strong differences in the fluorescence emission, which
indicates strong differences in DNA-binding properties (Fig. 3).
When looking at the differences between the different
hydrazide binding groups, one can note that no DNA com-
plexation occurs with AcHyd, regardless the nature of the
scaffold it is assembled onto. In contrast, DNA complexation is
detected with cationic binding groups. These results are in
line with our previous observations that A8·Ac does not
complex ctDNA while A8·L-Arg is an effective DNA-binding
cluster,18 and thus validates this microplate format assay.
Effect of multivalency. These results suggest a multivalent
effect since there are more hits with the tetravalent scaffold A8
than with the monovalent A1. The comparison of these results
with the DNA-binding ability of individual components con-
firms that multivalency is a major factor that account for DNA-
binding. Indeed, when tested alone by the EthBr assay, the
individual aldehyde scaffolds A were found unable to complex
ctDNA under these conditions (Fig. 4). Similarly, the individual
hydrazide binding groups show no DNA complexation under
these conditions, except for the tripeptides Lys3Hyd,
G1-LysHyd, and Arg3Hyd (Fig. 4).
Fig. 4 DNA complexation by individual aldehyde scaffolds and hydra-
zide binding groups determined by EthBr assay.
expressed by the N/P ratio of positive charges brought by the
clusters per negative charges in DNA. For instance, no com-
plexation was detected with L-ArgHyd in the range of N/P =
20–500 while the tetravalent cluster A8·L-Arg complexes
plasmid DNA effectively at N/P ≥ 20.18 Similarly, we found that
the tripeptide Arg3Hyd and the corresponding tetravalent
cluster A8·Arg3 complexes DNA at N/P ≥ 600 and N/P ≥ 1,
respectively (Fig. 5).
Effect of scaffolds. With the L-ArgHyd binding group, we
previously found that, when it is assembled onto the tetra-
valent scaffold A8, the corresponding cluster A8·L-Arg is an
effective DNA-binding agent.18 The results of the EthBr screen-
ing assay indicate that both the valency and the structure of
the scaffolds have a strong impact on the DNA-binding ability
of these clusters. Thus, divalent scaffolds A3, A4 and tri- and
tetra-valent scaffolds A6, A8 yield DNA-binding clusters. The
Gel retardation assays confirm the role of multivalency and
reveal the stoichiometry that is necessary for DNA binding –
This journal is © The Royal Society of Chemistry 2015
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
Fig. 8 Gel electrophoresis analyses showing the complexation of
plasmid DNA by clusters A8·L-Arg (left) and A7·L-Arg (right) at different
N/P.
Fig. 5 Gel electrophoresis analysis showing the complexation of
plasmid DNA by the tripeptide Arg3Hyd (left) and the corresponding
tetravalent cluster A8·Arg3 (right) at different N/P.
Fig. 6 Gel electrophoresis analysis showing the complexation of
plasmid DNA by clusters made of different scaffolds A with hydrazide
L-ArgHyd. Experiments carried out at N/P = 80.
Fig. 9 Gel electrophoresis analysis showing the complexation of
plasmid DNA by clusters made of scaffold A8 with different hydrazides.
Experiments carried out at N/P = 40.
While it remains unclear why the cyclic scaffold leads to a
more effective DNA complexation compared to the linear one,
this observation has been previously made with other cationic
clusters.12 In the present case, it is possible that the high pre-
organization offered by the constraint cyclic peptide scaffolds
A8 is at the origin of this effect by increasing the local concen-
tration of positive charges, thereby minimizing the entropic
penalty upon binding.
Effect of binding groups. The screening result shows that,
while neutral (AcHyd) or anionic hydrazide (AspHyd) give no
sign of DNA complexation when assembled on a scaffold, not
all cationic hydrazide give rise to DNA complexation when
Fig. 7 Gel electrophoresis analysis showing the complexation of
plasmid DNA by clusters made of different scaffolds A with hydrazide
Arg3Hyd. Experiments carried out at N/P = 1.
gel retardation assay confirms the importance of scaffold assembled on a scaffold. For instance, the clusters featuring
structure but tends to indicate a weaker binding of the divalent the Girard’s reagent (GirHyd) form poor DNA-binding agents
clusters compared to the tri- and tetra-valent ones (Fig. 6).
with most scaffolds even though it is positively charged. This
Similarly, with Arg3Hyd, the gel retardation assay carried indicates that, while electrostatic interactions may be impor-
out at N/P = 1 confirms the importance of the scaffold and tant for DNA complexation, other interactions such as hydro-
reveals the superiority of the same scaffolds: A3, A4, A6, and gen bonds must be considered as well. Gel retardation assay
A8 (Fig. 7).
confirms the general trend obtained by the EthBr screening
Along these studies, we noticed a strong difference of DNA- and indicates that, with the scaffold A8, amino acids and tri-
binding ability between the clusters derived from the linear peptides that contain lysine or arginine are the best binding
and cyclic scaffolds, respectively A7 and A8. According to groups (Fig. 9). A similar preference was observed with the
EthBr displacement assay, the cyclic cluster is more effective scaffold A6 (see ESI†).
than the linear one for DNA complexation. The gel retardation
The absence of DNA complexation observed by gel electro-
assay confirms the superiority of the cyclic cluster over the phoresis with amino acids of low pKa (His, Lys) is in contrast
linear one. Indeed, while the cyclic cluster bind effectively with the EthBr assay but is most likely explained by pH effects
plasmid DNA at N/P ≥ 20, the linear cluster is only effective at since the former was carried out at pH 8.0 while the latter was
N/P ≥ 50 (Fig. 8).
carried out at pH 5.0.24
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Organic & Biomolecular Chemistry
Paper
are not the sole responsible for the interactions between such
cationic biomolecular clusters and DNA.
Table 1 ctDNA complexation, assessed by fluorescence displacement
assay, by clusters A8·L-Arg and A8·Gly in different conditions
Overall, the fragments screening methodology enabled the
rapid and successful identification of clusters that effectively
bind DNA. The combination between the pre-organized
scaffold A8 and hydrazide binding groups featuring L-arginine(s)
leads to the most effective complexation of DNA.
[NaCl]
(mM)
Fluorescence
emissionb (%)
Entry
Compounds
N/Pa
1
2
3
4
5
6
A8·L-Arg
A8·L-Arg
A8·L-Arg
A8·Gly
A8·Gly
A8·Gly
80
2
80
80
2
9.4
9.4
150
9.4
9.4
150
2.6
0
25.3
3.6
100
92.4
80
Biological evaluation
a N/P represents the ratio of positive charges (N) per phosphodiester (P).
b Measured after 20 hours.
The potential of clusters A·Hyd as siRNA vectors was then eval-
uated on
a MCF7-luc cell line which is derived from
MCF7 human breast cancer cells transfected by firefly luci-
ferase gene. We first studied the cluster A8·L-Arg. Although
dynamic light scattering (DLS) experiments confirm complex
formation with siRNA and indicate a particle diameter of
148–190 nm at N/P = 20, no significant evidence of cell pene-
tration and transfection efficacy were observed by, respectively,
fluorescence imaging and cell luciferase assay (data not
shown). We then turned our attention to the more effective
cluster, A8·Arg3. Gel retardation assay confirms complex for-
mation with siRNA at N/P ≥ 9 (Fig. 11).
Similarly, DLS also shows complex formation with siRNA at
N/P = 9. The particles that are formed have diameter around
200 nm (PDI = 0.1–0.2) and show good stability over time. In
physiological salt concentration, the particles have a hydro-
dynamic diameter of 320 nm (PDI = 0.17) (Table 2).
Fig. 10 Gel electrophoresis analyses showing the complexation of
plasmid DNA by clusters A8·D-Arg (right) compared to A8·L-Arg (left) at
different N/P.
Then, the internalization of these particles by the cancer
cells was analysed by fluorescent microscopy. Satisfyingly, the
fluorescence imaging carried out with a Cy3-labeled non-
coding siRNA shows that the cluster A8·Arg3 effectively trans-
fects siRNA inside MCF7 cells (Fig. 12). This conclusion was
confirmed on another breast cancer cell line, MDA-MB-231
(see ESI†).
The screening results also show that the amino acids
bearing no positive charges on their side groups, such as
GlyHyd and AlaHyd, do not yield potent DNA binders when
assembled onto the tetravalent scaffold A8, thereby indicating
that the presentation of α amino groups is not sufficient for
promoting DNA binding. Nevertheless, when tested at N/P = 80
by the EthBr assay (pH 5.0), DNA complexation with cluster
A8·Gly can be detected. The comparison of DNA complexation
properties with cluster A8·L-Arg shows that, while the latter
does complex DNA effectively at low N/P = 2 and high salinity
(150 mM NaCl), the former only binds DNA at high N/P = 80
and low salinity (9.4 mM NaCl) (Table 1).
Gel retardation assays (pH 8.0) support this conclusion by
showing no DNA complexation with cluster A8·Gly in the range
of N/P = 10–80 while cluster A8·L-Arg clearly complexes DNA at
N/P ≥ 20 (see ESI†).18
Fig. 11 Gel electrophoresis analyses showing the complexation of
siRNA by cluster A8·Arg3 at different N/P.
We were intrigued by the apparent differences in DNA-
binding ability between clusters A8·L-Arg and A8·D-Arg (Fig. 9).
A cross-analysis by gel electrophoresis at different N/P con-
firms that, indeed, the chirality of the binding group has a
strong influence over the DNA-binding ability of the corres-
ponding clusters (Fig. 10). With cluster A8·D-Arg, the gel retar-
dation assay shows, in the range of N/P = 10–100, the
appearance of a band slightly above the band of plasmid DNA.
Compared to the results obtained with cluster A8·L-Arg, this
suggests a weaker DNA complexation and most probably indi-
cates different structures of the complexes with DNA. Again,
this result suggests that long-range electrostatic interactions
Table 2 Particle size characterization by dynamic light scattering of the
complex formed between cluster A8·Arg3 and siRNA at N/P = 9. Results
are expressed as mean standard deviation (n = 3)
After 1 hour
(154 mM NaCl)
t0
After 1 hour
Z-averagea
179.3 1.2
0.143 0.005
201.8 1.9
0.154 0.009
320.4 30.0
0.122 0.008
PDIb
a Hydrodynamic diameter in nm. b Polydispersity index.
This journal is © The Royal Society of Chemistry 2015
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
Experimental
Materials and methods
All solvents and chemical reagents were purchased from com-
mercial suppliers and used without further purifications. Dry
solvents were purchased in anhydrous quality from Fisher
Scientific (Acros). For dichloromethane, amylene was the
stabilizer. Calf-thymus DNA (ctDNA) was purchased from
Sigma Aldrich.
NMR. NMR spectra were recorded on Bruker Avance
400 instruments and were referenced with respect to the
residual solvent peak. Data are reported as follows:
chemical shift (δ in ppm), multiplicity (s for singlet,
d for doublet, t for triplet, q for quadruplet, m for multiplet,
bs for broad signal), coupling constant (J in Hertz) and
integration.
Mass spectrometry. MALDI-TOF and high resolution mass
spectrometry analyses were carried out at the Laboratoire de
Mesures Physiques, IBMM, Université de Montpellier, and
were obtained, respectively, on an Ultraflex III and a Waters
MicromassQ ToF mass spectrometer (positive mode).
HPLC. HPLC analyses were performed on a Waters HPLC
2695 (EC Nucleosil 300-5 C18, (125 × 3 mm) column,
Macherey-Nagel) equipped with a Waters 996 DAD detector.
The following linear gradients of solvent B (acetonitrile) into
solvent A (100 mM triethylammonium acetate/acetonitrile
95/5) were used: 0 to 95% of solvent B in 45 min. Retention
times (tR) are given in minutes. Semi-preparative RP-HPLC was
performed on a Waters 515 HPLC (VP Nucleodur HTec C18,
7 µm, (250 × 21) column, Macherey-Nagel) equipped with a
Waters 2487 detector. Preparative HPLC was performed on a
Waters Prep LC Controller HPLC (XSelect CSH Prep C18, 5 µm,
(250 × 30 mm) column, Macherey-Nagel) equipped with a
Waters 2489 detector.
Fig. 12 Fluorescence imaging (magnification 40×) of MCF7 cells, alone
(top), with Cy3-labeled non-coding siRNA (middle), and with the
complex formed between cluster A8·Arg3 and Cy3-labeled non-coding
siRNA at N/P = 9 (bottom). The blue fluorescence (left) indicates the
nuclei (DAPI staining), and red fluorescence (middle) indicates the Cy3-
labeled siRNA. Bars represent 50 μm.
The luciferase assay carried out with the 21-mer siRNA tar-
geting the expression of luciferase further shows that the
cluster A8·Arg3 acts as an effective vector to deliver siRNA
inside cells at N/P = 9. A dose-dependent silencing of the
expression of luciferase was observed by fluorescence
spectroscopy with increasing amounts of siRNA being trans-
fected (Fig. 13). The cell viability was assessed by a MTT
assay and shows no significant cell toxicity up to 0.5 μM of
cluster A8·Arg3 (Fig. 13). These results show that there is
a range of concentration (0.1–0.5 µM siRNA) where the cluster
A8·Arg3 shows good transfection efficacy and no cellular
toxicity.
SPPS. Solid-phase peptide synthesis (SPPS) was carried out
on a Liberty, CEM© instrument equipped with microwave.
The following conditions were used:
Solid support: 2-chlorotrityl chloride resin.
Coupling conditions: Fmoc-AA-OH (3 eq.), HBTU (3 eq.),
DIEA (5 eq.), DMF, 70 °C for 5 min with microwave irradiation
(0.25 mmol scale: 40 W, 0.1 mmol scale: 23 W). Double coup-
Conclusions
We reported herein a rapid screening methodology for deter- ling (7 min) was used for arginine. In order to prevent diketo-
mining the DNA-binding ability of various clusters formed by piperazine formation, the coupling of the amino acids next
acylhydrazone ligations. The results show that the structure of to Pro–Gly sequence was performed without heating or
the scaffold has a profound impact on the multivalent re- microwaves.
cognition of DNA. Furthermore, the detailed role of binding
Deprotection conditions: piperidine/DMF (2/8). 75 °C/40 W
group structure and chirality has been evidenced. The poten- for 30 s, then 70 °C/45 W for 3 min (twice).
tial of these clusters was then evaluated on cell culture and it
Cleavage conditions: TFA/CH2Cl2 (1/99), then MeOH/pyri-
was found that the most potent DNA-binding cluster also com- dine (8/2).
plexes, transports, and delivers effectively siRNA inside cells
Fluorescence displacement assay. Ethidium Bromide dis-
without displaying significant toxicity. The results pave the way placement assays were carried out either in cuvettes (3 mL
towards the application of these biomolecular clusters as artifi- quartz cuvette for experiments with N/P = 80, 0.5 mL quartz
cial vectors for the transport of oligonucleotides of therapeutic cuvette for experiments with N/P = 2) on a Hitashi Fluo-
interest. Further chemical modifications of the hits identified rescence Spectrophotometer F-2500, or in COSTAR 96-well
herein will be studied in order to optimize the transfection plates on a SAFAS Monaco Xenius XML instrument. Excitation
efficacy by, for instance, attaching lipophilic tails.25
wavelength was set at 546 nm and emission was measured
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Organic & Biomolecular Chemistry
Paper
scaffold. The fluorescence measurement was recorded after
shaking the solution a few seconds.
Experiments in 96-well plates were carried out as follows:
scaffolds A (20 mM) and hydrazide building blocks Hyd
(160 mM) were mixed in MilliQ H2O, in each well, and left
overnight at room temperature. Buffer, ctDNA, and EthBr were
then added (final concentration in scaffold: 1 mM). The plates
was shaken by orbital agitation (16 Hz, 120 s), and the
fluorescence emission was measured during 1 s using a 1200 V
voltage and a bandwidth of 6 nm for excitation and 10 nm for
emission.
Gel electrophoresis. Gel retardation assays with plasmid
DNA were performed using 100 ng of 5.7 kilobase pair
expression vector pET-15b (Novagen), which were mixed with
the appropriate amounts of ligand (in order to reach the
desired N/P ratio) in 0.5× TAE (20 mM Tris-acetate/0.5 mM
EDTA, pH 8.0) or 0.5× TBE (20 mM Tris-borate/0.5 mM EDTA,
pH 8.0) buffer to obtain a final volume of 10 µL. 2 µL of Blue
6× loading dye (Fisher Scientific) was added, after which 12 µL
was run on a 0.7% agarose gel (50 V). DNA was visualized with
SYBER Safe (Life Technologies) or SafeView Plus™ Nucleic
Acid Stain (Euromedex). The reference for the gel is GeneRuler
1 kb DNA ladder (250 to 10 000 bp) from Fischer Scientific. Gel
retardation assays with siRNA were performed as follows: the
cluster : siRNA complexes at a desired N/P were prepared in
RNase free water with 300 ng of eGFP siRNA. Electrophoresis
was carried out on a 1% wt vol−1 agarose gel in 1× TBE at 55 V
for 1 h. The GelRed-stained siRNA was visualized on an ultra-
violet transilluminator with a camera.
Dynamic light scattering. Particle size measurements were
carried out on a Zetasizer Nano ZS (Malvern, United Kingdom)
with transparent ZEN0040 disposable microcuvette cells
(40 µl) at 25 °C. The cluster : siRNA complexes were prepared
at a defined N/P ratio with a siRNA concentration of 1 µM.
Measurements were performed immediately after the com-
plexes formation and after 1 h incubation. Then, NaCl solution
was added at a final concentration of 154 mM and measure-
ment was repeated.
Fig. 13 Luciferase activity (top) and MTT cell viability (bottom) assays
showing, respectively, the transfection of a 21-mer siRNA targeting the
expression of luciferase inside MCF7-luc, and the viability of those cells.
The experiments were carried out with increasing amounts of siRNA (0.1
to 2 μM) contained in the A8·Arg3 : siRNA complex (N/P = 9). The results
are expressed as mean standard deviation (n = 3).
at 590 mm. The relative fluorescence emission decrease was
calculated as follows:
ꢀ
ꢁ
IEthBrꢀDNA ꢀ I
Fluorescence emission decrease ð%Þ ¼
Biological evaluation on cell culture. The MCF7-luc cell line
derived from MCF7 human breast cancer cells by stable trans-
IEthBrꢀDNA ꢀ IEthBr
where IEthBr–DNA represents the fluorescence emission of the fection of firefly luciferase gene (PCDNA 3.1 CMV-Luc-SVNeo)
EthBr–ctDNA intercalation complex, I represents the measured were generously provided by Dr P. Balaguer (ICM Montpellier,
fluorescence emission, and IEthBr represents the fluorescence France). Selection of resistant clones was performed by geneti-
emission of EthBr in the absence of ctDNA.
cin addition at 1 mg mL−1 until experiments. Cells were grown
The samples were prepared in the following buffers: (i) in phenol red-free F12/Dulbecco’s modified Eagle’s medium
sodium acetate (100 mM, pH 5.0), EDTA (10 μM), NaCl (DMEM) supplemented with 10% fetal calf serum (FCS). The
(9.4 mM), or (ii) Hepes buffer (100 mM, pH 7.2), EDTA cells were incubated at 37 °C in a humidified atmosphere
(10 μM), NaCl (9.4 mM), or (iii) sodium acetate (100 mM, pH with 5% CO2.
5.0), EDTA (10 μM), NaCl (150 mM).
Cy3-labeled non-coding siRNA was purchased from Euro-
Experiments in cuvette format were carried out as follows: gentec (Serring, Belgium). The siRNA targeting sequence for
ethidium bromide was dissolved in the buffer to provide luciferase (AACTTACGCTGAGTACTTCGA) was purchased from
42 μM concentration and the fluorescence was measured. The QIAGEN. The sense sequence is CUUACGCUGAGUACUUCGA
appropriate amount of ctDNA was added from a stock solution and the antisense sequence is UCGAAGUACUCAGCGUAAG.
in MilliQ H2O to reach the desired N/P. The ligands were then
The MTT (3-(4,5-dimethyl-thiazol-2-yl)-2,5-diphenyltetrazo-
added in microliters from 20–500 mM stock solutions in lium bromide, a yellow tetrazole) assay kit was purchased from
MilliQ H2O to provide a final concentration of 1 mM in Sigma-Aldrich (Saint-Quentin-Fallavier, France).
This journal is © The Royal Society of Chemistry 2015
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
The cluster : siRNA complexes were prepared as follows: the follows: 10 mg of resin were suspended in 1 mL of piperidine/
clusters A·Hyd (20 mM in filtered RNase-free water) were DMF (2/8) for 20 min and filtered. 100 µL of the filtrate were
vortexed for 5 s, ultrasonically mixed for 5 min, mixed with dissolved in 10 mL of DMF and the absorbance was measured
siRNA in order to reach the desired N/P, and incubated at at 301 nm. The charge of the modified resin was determined
room temperature for 20 min.
to be 0.31 mmol g−1, and it was then engaged in SPPS. The
Fluorescence microscopy. MCF7-luc cells were seeded at a final acetylation was then carried out with 10 mL of a solution
density of 50 000 cells per well and incubated for 24 h in a of acetic anhydride/CH2Cl2 (50/50) for 20 min (2×). The
8-well tissue culture chambers (Sarstedt, Germany) before product was then cleaved and deprotected with a TFA/H2O/TIS
treatment. The cells were incubated with A8·Arg3-Cy3-labeled (95/2.5/2.5) solution (20 mL) for 2 hours. The filtrate was
siRNA complexes at a molar ratio of 25/1 (N/P = 9) in OptiMEM concentrated, precipitated by adding Et2O, and centrifuged
(Invitrogen) for 4 h. These cells were then fixed with 500 μl of (10 000 t min−1 for 10 min). The supernatant was removed and
cold 4% PFA for 30 min and washed twice with PBS. The the crude material was freeze dried to give a white solid
nuclei were stained using a DAPI solution (Sigma-Aldrich, (326 mg, 71%). The oxidative cleavage was carried out by dis-
Saint Quentin-Fallavier, France). Images were taken with an solving the crude material in 18.4 mL of H2O and adding
inverted fluorescence microscope (Leica DMRB, Leica Micro- 403 mg (1.8 mmol, 10 eq.) of sodium periodate. The desired
systems GmbH, Germany) and analyzed using ImageJ product was then purified by semi-preparative HPLC using the
software.
following gradient of ACN/TFA 99.9/0.1 into H2O: 0 to 25% in
Cell luciferase assay. MCF7-luc cells were seeded at a 25 min, then 25% to 100% in 10 min. 104 mg (47%) of a white
density of 5000 cells per well in 96-well white opaque tissue solid were obtained (overall yield for the preparation of A7:
culture plates in 150 μl complete culture medium and incu- 51%). HPLC: tR 3.49 min; MALDI-ToF (HCCA): calcd for
bated for 24 h. The cells were then washed with PBS and incu- [C54H82N14O20 + K]+ 1285.58, found 1285.52.
bated with various cluster : siRNA complex formulations in
General procedure for the solid-phase synthesis of amino
OptiMEM at 37 °C for 4 h. Thereafter, 50 μl of 40% serum con- acid hydrazides. Preparation of the modified resin: DIEA
taining medium was added. Two days after transfection, (2.45 mL, 14.05 mmol) was added to a suspension of 2-chloro-
expression of luciferase was assessed by addition into culture trityl chloride resin (1.75 g, 2.81 mmol Cl g−1) in NMP/DMSO
medium of luciferin (10−3 M, final concentration) purchased (4/1) (9 mL). A solution of 9-fluorenylmethyl carbazate (Fmoc-
from Promega (France). Living cell luminescence was hydrazide, 2.14 g, 8.43 mmol) in DMSO (9 mL) was then added
measured 10 min after by
a multilabel plate reader and the mixture was stirred for 2 hours. After filtration,
(Wallac1420, PerkinElmer, USA) for 5 s. Values are expressed capping was performed by mixing with MeOH for a few
as percentage of luciferase activity compared to non-treated minutes. The resin was then filtered, washed with DMSO (3×),
wells (set as 100%).
CH2Cl2 (2×), isopropanol (1×), CH2Cl2 (2×) and Et2O (3×), and
Cytotoxicity assay. The cytotoxicity of cluster : siRNA com- dried in vacuo. The loading, calculated according to the pro-
plexes was determined by the MTT assay. In brief, MCF7-luc cedure used in the synthesis of A7, was determined to be
cells were seeded at 5000 cells per well in clear, flat-bottomed, 0.39 mmol g−1
.
96-well plates (Costar) 24 h before treatment. After being
Peptide couplings procedure: the initial Fmoc deprotection
washed, 150 μl of OptiMEM that contained the cluster : siRNA was carried out as follows: the resin was suspended in a solu-
complexes at different N/P ratios was added to the wells and tion of piperidine/NMP/DMSO (30/56/14) (2 mL). After 1 h, it
incubated for 4 h. Thereafter, 50 μl of 40% serum containing was filtered, washed with DMSO (3×), CH2Cl2 (2×), Isopropanol
medium was added, and the cytotoxicity of the relevant (1×), CH2Cl2 (2×), Et2O (3×), and dried in vacuo. The resin
reagents was determined by the MTT assay after 48 h. The (500 mg, 0.195 mmol) was suspended in a solution of NMP/
absorbance at 540 nm was read on a plate reader (Multiskan DMSO (4/1) (500 µL). The Fmoc-amino acid-OH (0.351 mmol)
FC, Thermo Fisher Scientific, France). The results obtained was pre-activated by addition, in a solution of NMP/DMSO
from triplicate wells were averaged and normalized to the (4/1) (1 mL), of DIEA (98 µL, 0.585 mmol) and HATU
value obtained from the non-treated cells.
(133.5 mg, 0.351 mmol). After 30 min, the mixture of pre-acti-
vated amino acid was added to the resin. After 3 hours, the
resin was filtered and this coupling procedure was repeated
Synthetic procedures
Linear peptide A7. Similarly to the synthesis of cyclic twice. The resin was then washed (DMSO (3×), CH2Cl2 (2×), Iso-
peptide A8,18 the decapeptide was synthesized via SPPS strat- propanol (1×), CH2Cl2 (2×) and Et2O (3×)) and dried. Capping
egy (0.25 mmol scale) using 2-chlorotrityl chloride resin and a was carried out by suspending the resin in a solution of acetic
Fmoc-Lys[Boc-Ser(t-Bu)]-OH26 dipeptide. The resin was loaded anhydride/CH2Cl2 (1/1) (2 mL) for 20 min (procedure repeated
with the first amino acid by reacting 3 g (4.8 mmol Cl−1) of twice). The resin was then filtered, washed (DMSO (3×),
resin with 4.3 g (14.4 mmol, 3 eq.) of Fmoc-Gly-OH and 4.2 mL CH2Cl2 (2×), Isopropanol (1×), CH2Cl2 (2×) and Et2O (3×)) and
(24 mmol, 2 M, 5 eq.) of DIEA in 40 mL DMF/dichloromethane dried in vacuo. Fmoc deprotection was carried out by suspend-
for 1.5 hour. The resin was then filtered, washed with MeOH, ing the resin in a solution of piperidine/NMP/DMSO (30/56/14)
DMF (3×), CH2Cl2 (2×), diethyl ether (1×) and dried in vacuo. (2 mL) for 1 hour. The resin was then filtered, washed and
The loading was quantified by UV-Vis spectroscopy27 as dried in vacuo.
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Organic & Biomolecular Chemistry
Paper
1
Deprotection and cleavage procedure: the resin was sus- 117.8 (q, J(C,F) = 290, 1C, F3C–), 51.6, 40.3, 27.8, 23.5; [α]2D0:
pended in a solution of TFA/TIS/H2O (95/2.5/2.5) (5 mL) for −7.2 (H2O, c = 1.0); HR-ESI-MS: calcd for [C6H16N6O + H]+
2 hours, then filtered and washed with CH2Cl2. The filtrate 189.1464, found 189.1464.
was partially concentrated in vacuo, and precipitated with
Et2O. After centrifugation and removal of the supernatant, the 1-oxohexan-2-yl)amino)-1-oxohexan-2-yl)hexanamide (Lys3Hyd).
white solid was freeze dried. Lys3Hyd was synthesized by SPPS on a 0.1 mmol scale using
(S)-2,6-Diamino-N-((S)-6-amino-1-(((S)-6-amino-1-hydrazinyl-
2-Aminoacetohydrazide (GlyHyd). GlyHyd was synthesized the modified Fmoc hydrazine resin that is manually prepared
1
according general procedure (60.6 mg, quantitative). H NMR from the 2-chlorotrityl chloride resin as described above. After
(D2O; 400 MHz) δ 3.88 (s, 2H, H2N–CH2); 13C NMR (D2O; SPPS, the product was cleaved from the resin with a solution
2
100 MHz) δ 166.0, 163.1 (q, J(C,F) = 35, 1C, F3C–CvO), 117.8 of TFA/CH2Cl2 (1 : 99) (10 mL) (3 min, 3×). A solution of
(q, 1J(C,F) = 290, 1C, F3C–), 39.2.
MeOH/pyridine (8/2) (10 mL) was added to the filtrate and the
(S)-2-Aminopropanehydrazide (AlaHyd). AlaHyd was syn- solution was partially concentrated in vacuo. Et2O was added
thesized according general procedure (21.2 mg, 67%). 1H NMR to the crude product and a white precipitate appeared. The
2
2
(D2O; 400 MHz) δ 4.19 (q, J = 7.0, 1H, H2N–CH), 1.57 (d, J = mixture was centrifuged, the supernatant was removed and the
7.0, 3H, CH–CH3); 13C NMR (D2O; 100 MHz) δ 169.2, 163.1 (q, precipitate was freeze dried. The crude product was then puri-
1
2J(C,F) = 35, 1C, F3C–CvO), 117.8 (q, J(C,F) = 290, 1C, F3C–), fied on preparative HPLC using the following gradient of ACN/
47.9, 16.2; [α]2D0: +8.1 (H2O, c = 1.0); HR-ESI-MS: calcd for TFA 99.9/0.1 into H2O: 20% to 40% in 20 min to obtain a
[C3H9N3O + H]+ 104.0824, found 104.0822.
white solid after freeze-drying. Final deprotection was carried
(S)-3-Amino-4-hydrazinyl-4-oxobutanoic
acid
(AspHyd). out with TFA/H2O/TIS (95/2.5/2.5) (10 mL) for 2 hours. The
AspHyd was synthesized according general procedure deprotected product was partially concentrated in vacuo. Et2O
1
(19.2 mg, 53%). H NMR (D2O; 400 MHz) δ 4.43 (m, 1H, CH– was then added and a white precipitate appeared. The mixture
CH2), 3.08 (m, 2H, CH–CH2–CO); 13C NMR (D2O; 100 MHz) was centrifuged and the precipitate dried. Lys3Hyd was freeze-
2
1
δ 172.3, 167.3, 163.1 (q, J(C,F) = 35, 1C, F3C–CvO), 117.8 (q, dried and obtained as a white solid (49.3 mg, 52%). H NMR
1J(C,F) = 290, 1C, F3C–), 48.5, 34.6; [α]2D0: +17.4 (H2O, c = 1.0); (D2O; 400 MHz) δ 4.38–4.35 (m, 2H, 2CH–CH2), 4.07 (t, 3J = 6.0,
HR-ESI-MS: calcd for [C4H9N3O3 + H]+ 148.0722, found 1H, CH–CH2), 3.05–3.04 (m, 6H, 3CH2–NH2), 1.95–1.72 (m,
148.0720.
12H, 3CH–CH2, 3CH2–CH2–NH2), 1.55–1.45 (m, 6H, 3CH–CH2–
(S)-2-Amino-3-(1H-imidazol-5-yl)propanehydrazide (HisHyd). CH2); 13C NMR (D2O; 100 MHz) δ 173.6, 171.8, 169.7, 163.1 (q,
HisHyd was synthesized according general procedure 2J(C,F) = 35, 1C, F3C–CvO), 117.8 (q, J(C,F) = 290, 1C, F3C–),
1
1
(26.1 mg, 68%). H NMR (D2O; 400 MHz) δ 8.72 (s, 1H, HN– 53.7, 52.8, 52.2, 39.2, 39.0, 30.6, 30.5, 30.3, 26.4, 26.3, 22.1,
CHvN), 7.46 (s, 1H, CvCH–N), 4.36 (t, J = 7.0, 1H, CH–CH2); 22.0, 21.3; [α]2D0: −11.6 (H2O, c = 1.0); HR-ESI-MS: calcd for
3
13C NMR (D2O; 100 MHz) δ 166.6, 163.1 (q, 2J(C,F) = 35, 1C, [C18H40N8O3 + H]+ 417.3302, found 417.3296.
F3C–CvO), 135.6, 125.4, 118.6, 117.8 (q, 1J(C,F) = 290, 1C,
(2S,2′S)-N,N′-((S)-6-Hydrazinyl-6-oxohexane-1,5-diyl)bis(2,6-
F3C–), 50.9, 25.8; [α]D20: +18.0 (H2O, c = 1.0); HR-ESI-MS: calcd diaminohexanamide) (G1-LysHyd). G1-LysHyd was synthesized
for [C6H11N5O + H]+ 170.1042, found 170.1041.
according to the general procedure for the solid-phase syn-
(S)-2,6-Diaminohexanehydrazide (LysHyd). LysHyd was syn- thesis of amino acid hydrazides using Fmoc-L-Lys(Fmoc)-OH.
thesized according general procedure (37.1 mg, 84%). 1H NMR The second coupling reaction was carried out with doubled
2
2
(D2O; 400 MHz) δ 4.08 (t, J = 7.0, 1H, H2N–CH), 2.99 (t, J = quantities of reagents. Finally, the resin was suspended in a
7.6, 2H, CH2–NH2), 1.94 (m, 2H, CH2–CH2–NH2), 1.70 (m, 2H, solution of TFA/TIS/H2O (95/2.5/2.5) (5 mL) for 2 hours, fil-
2
CH–CH2); 13C NMR (D2O; 100 MHz) δ 168.2, 163.1 (q, J(C,F) = tered and washed with CH2Cl2. The filtrate was partially con-
35, 1C, F3C–CvO), 117.8 (q, 1J(C,F) = 290, 1C, F3C–), 51.8, centrated in vacuo, and precipitated with Et2O. After
39.0, 30.1, 26.3, 21.2; [α]2D0: +16.8 (H2O, c = 1.0); HR-ESI-MS: centrifugation and removal of the supernatant, the white solid
calcd for [C6H16N4O + H]+ 161.1402, found 161.1402.
was freeze dried to afford G1-LysHyd as a white solid
(S)-1-(4-Amino-5-hydrazinyl-5-oxopentyl)guanidine (L-ArgHyd). (230.8 mg, 84%). 1H NMR (MeOD; 400 MHz) δ 4.40 (bs, 1H,
L-ArgHyd was synthesized according general procedure CH–CH2), 3.96 (bs, 1H, CH–CH2), 3.85 (bs, 1H, CH–CH2), 3.25
(14.2 mg, 81%). 1H NMR (D2O; 400 MHz) δ 4.10 (t, 3J = 6.7, 1H, (bs, 2H, CH2–NH), 2.95 (bs, 4H, 2CH2–NH), 1.89–1.47 (m, 18H,
CH–CH2), 3.24 (t, 3J = 6.8, 2H, CH2–CH2–NH), 2.02–1.92 (m, 3CH–CH2–CH2, 3CH2–CH2–CH2, 3CH2–NH2); 13C NMR (MeOD;
2H, CH2–CH2–NH), 1.72–1.64 (m, 2H, CH–CH2–CH2); 13C NMR 100 MHz) δ 170.2, 170.0, 54.3, 53.9, 53.6, 40.3, 40.3, 32.3, 32.1,
(D2O; 100 MHz) δ 168.2, 156.8, 51.6, 40.2, 27.7, 23.5; [α]2D0: 31.9, 29.8, 28.1, 28.0, 24.1, 22.9, 22.5; [α]D20: +8.8 (H2O, c = 1.0);
+15.3 (H2O, c = 1.0); HR-ESI-MS: calcd for [C6H16N6O + H]+ HR-ESI-MS: calcd for [C18H40N8O3 + H]+ 417.3302, found
189.1464, found 189.1466.
417.3299.
(R)-1-(4-Amino-5-hydrazinyl-5-oxopentyl)guanidine (D-ArgHyd).
(S)-2-Amino-5-guanidino-N-((S)-5-guanidino-1-(((S)-5-guanidino-
D-ArgHyd was synthesized according general procedure 1-hydrazinyl-1-oxopentan-2-yl)amino)-1-oxopentan-2-yl)pentan-
(67.6 mg, 75%). 1H NMR (D2O; 400 MHz) δ 4.15 (t, 3J = 6.6, 1H, amide (Arg3Hyd). Arg3Hyd was synthesized by SPPS on a
3
CHCH2), 3.24 (t, J = 6.8, 2H, CH2–NH), 2.02–1.95 (m, 2H, CH– 0.1 mmol scale using the modified Fmoc hydrazine resin that
CH2), 1.73–1.65 (m, 2H, CH2CH2NH); 13C NMR (D2O; is manually prepared from the 2-chlorotrityl chloride resin as
2
100 MHz) δ 168.1, 163.1 (q, J(C,F) = 35, 1C, F3C–CvO), 156.9, described above. After SPPS, the product was cleaved from the
This journal is © The Royal Society of Chemistry 2015
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
resin with a solution of TFA/CH2Cl2 (1 : 99) (10 mL) (3 min,
3×). A solution of MeOH/pyridine (8/2) (10 mL) was added to
Acknowledgements
the filtrate and the solution was partially concentrated We thank the CNRS, the LabEx CheMISyst (ANR-10-LABX-05-
in vacuo. Et2O was added to the crude product and a white pre- 01) and the ANR (ANR-11-PDOC-002-02) for funding. Auto-
cipitate appeared. The mixture was centrifuged, the super- mated peptide synthesis was carried out at the SynBio 3 plat-
natant was removed and the precipitate was freeze dried. The form – IBMM, Montpellier. We thank the ENSCM for the
crude product was then purified on preparative HPLC using access to the fluorescence microplate reader. We thank Sarah
the following gradient of ACN/TFA 99.9/0.1 into H2O: 20% to Michel for her contribution to the peptide synthesis.
70% in 45 min to afford a white solid after freeze-drying. Final
deprotection was carried out with TFA/H2O/TIS (95/2.5/2.5)
(10 mL) for 2 hours. The deprotected product was partially
concentrated in vacuo and precipitated with Et2O. The mixture
References
was centrifuged and the precipitate dried. Arg3Hyd was freeze-
dried and obtained as a white solid (48.8 mg, 46%). H NMR
1 S. T. Crooke, Annu. Rev. Med., 2004, 55, 61–95;
A. Demesmaeker, R. Haner, P. Martin and H. E. Moser, Acc.
Chem. Res., 1995, 28, 366–374; E. Uhlmann and A. Peyman,
Chem. Rev., 1990, 90, 543–584.
2 J. C. Burnett, J. J. Rossi and K. Tiemann, Biotechnol. J.,
2011, 6, 1130–1146; J. W. Gaynor, B. J. Campbell and
R. Cosstick, Chem. Soc. Rev., 2010, 39, 4169–4184;
D. Castanotto and J. J. Rossi, Nature, 2009, 457, 426–
433.
1
(D2O; 400 MHz) δ 4.43–4.37 (m, 2H, 2CH–CH2), 4.12 (t, 3J = 6.4,
1H, CH–CH2), 3.26–3.23 (m, 6H, 3CH2–NH), 1.99–1.80 (m, 6H,
3CHCH2), 1.76–1.61 (m, 6H, 3CHCH2CH2); 13C NMR (D2O;
100 MHz) δ 173.6, 171.6, 169.7, 156.9, 53.7, 52.6, 52.0, 40.6,
40.4, 28.1, 28.0, 27.9, 24.4, 23.5; [α]2D0: −5.6 (H2O, c = 1.0);
HR-ESI-MS: calcd for [C18H40N14O3 + H]+ 501.3486, found
501.3482.
General procedure for clusters synthesis through acylhydra-
zone ligation. 8 equivalents of hydrazide per scaffold were
added from a 200 mM stock solution in water into a 1 mM
solution of scaffold A in aqueous buffer (100 mM AcONa, pH
5.0) and the mixture was stirred overnight at room tempera-
ture. The resulting cluster was analyzed by HPLC and MALDI-
ToF or HR-ESI mass spectrometry. Clusters A8·Ac and A8·L-Arg
were prepared as previously described.18
3 M. A. Mintzer and E. E. Simanek, Chem. Rev., 2009, 109,
259–302; S. Lehrman, Nature, 1999, 401, 517–518.
4 U. Lächelt and E. Wagner, Chem. Rev., 2015, DOI: 10.1021/
cr5006793.
5 S. K. Samal, M. Dash, S. V. Vlierberghe, D. L. Kaplan,
E. Chiellini, C. v. Blitterswijk, L. Moroni and P. Dubruel,
Chem. Soc. Rev., 2012, 41, 7147–7194; D. N. Nguyen,
J. J. Green, J. M. Chan, R. Longer and D. G. Anderson, Adv.
Mater., 2009, 21, 847–867; S. Y. Wong, J. M. Pelet and
D. Putnam, Prog. Polym. Sci., 2007, 32, 799–837; T. Merdan,
J. Kopecek and T. Kissel, Adv. Drug Delivery Rev., 2002, 54,
715–758.
A8·Gly MALDI-ToF (HCCA): calcd for [C60H98N26O18 + Na]+
1493.76, found 1493.72. A8·Gir HPLC: tR 6.83 min; MALDI-ToF
(HCCA): calcd for [C72H127N26O18 + 4Cl + Na]+ 1805.98, found
1805.86. A8·Ala MALDI-ToF (HCCA): calcd for [C64H106N26O18
+
Na]+ 1549.82, found 1549.81. A8·Asp MALDI-ToF (HCCA): calcd
for [C68H107N26O26 + Na]+ 1726.78, found 1725.76. A8·His
MALDI-ToF (HCCA): calcd for [C76H115N34O18 + Na]+ 1814.91,
found 1813.92. A8·Lys MALDI-ToF (HCCA): calcd for
6 X. X. Liu, P. Rocchi and L. Peng, New J. Chem., 2012, 36,
256–263.
7 C. Fasting, C. A. Schalley, M. Weber, O. Seitz, S. Hecht,
B. Koksch, J. Dernedde, C. Graf, E. W. Knapp and R. Haag,
Angew. Chem., Int. Ed., 2012, 51, 10472–10498;
G. V. Oshovsky, D. N. Reinhoudt and W. Verboom, Angew.
Chem., Int. Ed., 2007, 46, 2366–2393; L. L. Kiessling,
J. E. Gestwicki and L. E. Strong, Angew. Chem., Int. Ed.,
2006, 45, 2348–2368; J. D. Badjic, A. Nelson, S. J. Cantrill,
W. B. Turnbull and J. F. Stoddart, Acc. Chem. Res., 2005, 38,
723–732; A. Mulder, J. Huskens and D. N. Reinhoudt, Org.
Biomol. Chem., 2004, 2, 3409–3424; M. Mammen, S. K. Choi
and G. M. Whitesides, Angew. Chem., Int. Ed., 1998, 37,
2755–2794.
8 S. M. Moghimi, P. Symonds, J. C. Murray, A. C. Hunter,
G. Debska and A. Szewczyk, Mol. Ther., 2005, 11, 990–995.
9 For examples of disulfide-based bioreducible polymers and
dendrimers, see for instance: A. Barnard, P. Posocco,
M. Fermeglia, A. Tschiche, M. Calderon, S. Pricl and
D. K. Smith, Org. Biomol. Chem., 2014, 12, 446–455;
L. D. Feng, A. Xie, X. Y. Hu, Y. Y. Liu, J. F. Zhang, S. F. Li
and W. Dong, New J. Chem., 2014, 38, 5207–5214;
K. L. Kozielski, S. Y. Tzeng and J. J. Green, Chem. Commun.,
[C76H134N30O18
+
Na]+ 1778.05, found 1777.99. A8·D-Arg
MALDI-ToF (HCCA): calcd for [C76H134N38O18 + H]+ 1868.07,
found 1868.05. A8·Lys3 HPLC: tR 2.94 min; MALDI-ToF
(HCCA): calcd for [C124H231N46O26 + Na]+ 2803.82, found
2802.76. A8·G1-Lys HPLC: tR 4.66 min; MALDI-ToF (HCCA):
calcd for [C124H231N46O26 + Na]+ 2803.82, found 2802.79.
A8·Arg3 HPLC: tR 5.79 min; MALDI-ToF (HCCA): calcd for
[C124H231N70O26 + H]+ 3116.82, found 3116.86. A1·Ac HPLC:
tR 6.26 min; HR-ESI-MS: calcd for [C9H10N2O + Na]+ 185.0871,
found 185.0870. A2·Ac HPLC: tR 7.40 min; HR-ESI-MS: calcd
for [C10H10N2O2 + Na]+ 213.0821, found 213.0819. A3·Ac HPLC:
tR 6.78 min; HR-ESI-MS: calcd for [C12H14N4O2
+
Na]+
269.1195, found 269.1197. A4·Ac HPLC: tR 6.92 min;
HR-ESI-MS: calcd for [C12H14N4O2 + Na]+ 269.1195, found
269.1197. A5·Ac HR-ESI-MS: calcd for [C9H16N4O2 + Na]+
235.1171, found 235.1169. A6·Ac HPLC: tR 7.92 min;
HR-ESI-MS: calcd for [C15H18N6O3 + Na]+ 353.1519, found
353.1516. A7·Ac HPLC: tR 7.99 min; MALDI-ToF (HCCA): calcd
for [C62H98N22O20 + Na]+ 1493.73, found 1493.70.
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Organic & Biomolecular Chemistry
Paper
2013, 49, 5319–5321; H. M. Liu, H. Wang, W. J. Yang and 15 V. Villari, A. Mazzaglia, R. Darcy, C. M. O’Driscoll and
Y. Y. Cheng, J. Am. Chem. Soc., 2012, 134, 17680–17687;
S. Son, R. Namgung, J. Kim, K. Singha and W. J. Kim, Acc.
Chem. Res., 2012, 45, 1100–1112; Y. Wang, R. Zhang, N. Xu,
F. S. Du, Y. L. Wang, Y. X. Tan, S. P. Ji, D. H. Liang and
Z. C. Li, Biomacromolecules, 2011, 12, 66–74; T. I. Kim,
M. Lee and S. W. Kim, Biomaterials, 2010, 31, 1798–1804;
J. G. Hardy, C. S. Love, N. P. Gabrielson, D. W. Pack and
D. K. Smith, Org. Biomol. Chem., 2009, 7, 789–793;
S. Bauhuber, C. Hozsa, M. Breunig and A. Gopferich, Adv.
Mater., 2009, 21, 3286–3306.
N. Micali, Biomacromolecules, 2013, 14, 811–817;
A. Martinez, C. Bienvenu, J. L. J. Blanco, P. Vierling,
C. O. Mellett, J. M. G. Fernandez and C. Di Giorgio, J. Org.
Chem., 2013, 78, 8143–8148; C. Bienvenu, A. Martinez,
J. L. J. Blanco, C. Di Giorgio, P. Vierling, C. O. Mellet,
J. Defaye and J. M. G. Fernandez, Org. Biomol. Chem., 2012,
10, 5570–5581; A. Mendez-Ardoy, N. Guilloteau, C. Di
Giorgio, P. Vierling, F. Santoyo-Gonzalez, C. O. Mellet and
J. M. G. Fernandez, J. Org. Chem., 2011, 76, 5882–5894;
A. Diaz-Moscoso, L. Le Gourrierec, M. Gomez-Garcia,
J. M. Benito, P. Balbuena, F. Ortega-Caballero,
N. Guilloteau, C. Di Giorgio, P. Vierling, J. Defaye,
C. O. Mellet and J. M. G. Fernandez, Chem. – Eur. J., 2009,
15, 12871–12888; A. Diaz-Moscoso, P. Balbuena, M. Gomez-
Garcia, C. O. Mellet, J. M. Benito, L. Le Gourrierec, C. Di
Giorgio, P. Vierling, A. Mazzaglia, N. Micali, J. Defaye and
J. M. G. Fernandez, Chem. Commun., 2008, 2001–2003;
F. Ortega-Caballero, C. O. Mellet, L. Le Gourrierec,
N. Guilloteau, C. Di Giorgio, P. Vierling, J. Defaye and
J. M. G. Fernandez, Org. Lett., 2008, 10, 5143–5146.
10 For examples of acid-sensitive materials, see for instance:
C. Bouillon, D. Paolantoni, J. C. Rote, Y. Bessin,
L. W. Peterson, P. Dumy and S. Ulrich, Chem. – Eur. J.,
2014, 20, 14705–14714; H. L. Cao, Y. X. Dong, A. Aied,
T. Y. Zhao, X. Chen, W. X. Wang and A. Pandit, Chem.
Commun., 2014, 50, 15565–15568; S. Binauld and
M. H. Stenzel, Chem. Commun., 2013, 49, 2082–2102;
B. J. Hong, A. J. Chipre and S. T. Nguyen, J. Am. Chem. Soc.,
2013, 135, 17655–17658; S. Duan, W. Yuan, F. Wu and
T. Jin, Angew. Chem., Int. Ed., 2012, 51, 7938–7941;
L. N. Cui, J. L. Cohen, C. K. Chu, P. R. Wich, 16 X. Zhang, Z. Zhang, X. Xu, Y. Li, Y. Li, Y. Jian and Z. Gu,
P. H. Kierstead and J. M. J. Frechet, J. Am. Chem. Soc., 2012,
134, 15840–15848; A. Barnard, P. Posocco, 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; J. A. Wolff and D. B. Rozema, Mol. Ther.,
2008, 16, 8–15; D. B. Rozema, D. L. Lewis, D. H. Wakefield,
S. C. Wong, J. J. Klein, P. L. Roesch, S. L. Bertin,
T. W. Reppen, Q. Chu, A. V. Blokhin, J. E. Hagstrom and
J. A. Wolff, Proc. Natl. Acad. Sci. U. S. A., 2007, 104, 12982–
12987; A. Aissaoui, B. Martin, E. Kan, N. Oudrhiri,
M. Hauchecorne, J. P. Vigneron, J.-M. Lehn and P. Lehn,
J. Med. Chem., 2004, 47, 5210–5223.
Angew. Chem., Int. Ed., 2015, 54, 4289–4294; X. X. Liu,
J. H. Zhou, T. Z. Yu, C. Chen, Q. Cheng, K. Sengupta,
Y. Y. Huang, H. T. Li, C. Liu, Y. Wang, P. Posocco,
M. H. Wang, Q. Cui, S. Giorgio, M. Fermeglia, F. Q. Qu,
S. Pricl, Y. H. Shi, Z. C. Liang, P. Rocchi, J. J. Rossi and
L. Peng, Angew. Chem., Int. Ed., 2014, 53, 11822–11827;
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. Pharm., 2011, 8, 416–429; D. J. Welsh,
S. P. Jones and D. K. Smith, Angew. Chem., Int. Ed., 2009,
48, 4047–4051; S. P. Jones, N. P. Gabrielson, D. W. Pack and
D. K. Smith, Chem. Commun., 2008, 4700–4702;
M. A. Kostiainen, J. G. Hardy and D. K. Smith, Angew.
Chem., Int. Ed., 2005, 44, 2556–2559.
11 P. L. Padnya, E. A. Andreyko, O. A. Mostovaya,
I. K. Rizvanov and I. I. Stoikov, Org. Biomol. Chem., 2015,
13, 5894–5904; V. Bagnacani, V. Franceschi, L. Fantuzzi, 17 G. M. Pavan, M. A. Kostiainen and A. Danani, RSC Adv.,
A. Casnati, G. Donofrio, F. Sansone and R. Ungaro, Bio-
conjugate Chem., 2012, 23, 993–1002; F. Sansone, L. Baldini,
A. Casnati and R. Ungaro, New J. Chem., 2010, 34, 2715–
2728; V. Bagnacani, F. Sansone, G. Donofrio, L. Baldini,
2011, 1, 1677–1681; P. Posocco, S. Pricl, S. Jones,
A. Barnard and D. K. Smith, Chem. Sci., 2010, 1, 393–404;
G. M. Pavan, A. Danani, S. Pricl and D. K. Smith, J. Am.
Chem. Soc., 2009, 131, 9686–9694.
A. Casnati and R. Ungaro, Org. Lett., 2008, 10, 3953–3956; 18 E. Bartolami, Y. Bessin, V. Gervais, P. Dumy and S. Ulrich,
F. Sansone, M. Dudic, G. Donofrio, C. Rivetti, L. Baldini, Angew. Chem., Int. Ed., 2015, DOI: 10.1002/anie.201504047.
A. Casnati, S. Cellai and R. Ungaro, J. Am. Chem. Soc., 2006, 19 C. Gehin, J. Montenegro, E.-K. Bang, A. Cajaraville,
128, 14528–14536.
S. Takayama, H. Hirose, S. Futaki, S. Matile and
H. Riezman, J. Am. Chem. Soc., 2013, 135, 9295–9298;
J. Montenegro, E. K. Bang, N. Sakai and S. Matile, Chem. –
Eur. J., 2012, 18, 10436–10443; J. Montenegro, A. Fin and
S. Matile, Org. Biomol. Chem., 2011, 9, 2641–2647;
J. Montenegro, P. Bonvin, T. Takeuchi and S. Matile, Chem.
– Eur. J., 2010, 16, 14159–14166.
12 V. Bagnacani, V. Franceschi, M. Bassi, M. Lomazzi,
G. Donofrio, F. Sansone, A. Casnati and R. Ungaro, Nat.
Commun., 2013, 4, 1721.
13 D. Sigwalt, M. Holler, J. Iehl, J. F. Nierengarten,
M. Nothisen, E. Morin and J. S. Remy, Chem. Commun.,
2011, 47, 4640–4642.
14 Y. C. Chang, K. Yang, P. Wei, S. S. Huang, Y. X. Pei, 20 Y. C. Huang, C. C. Chen, S. J. Li, S. Gao, J. Shi and Y. M. Li,
W. Zhao and Z. C. Pei, Angew. Chem., Int. Ed., 2014, 53, Tetrahedron, 2014, 70, 2951–2955.
13126–13130; I. Nierengarten, M. Nothisen, D. Sigwalt, 21 D. L. Boger, B. E. Fink, S. R. Brunette, W. C. Tse and
T. Biellmann, M. Holler, J. S. Remy and J. F. Nierengarten,
M. P. Hedrick, J. Am. Chem. Soc., 2001, 123, 5878–
Chem. – Eur. J., 2013, 19, 17552–17558.
5891.
This journal is © The Royal Society of Chemistry 2015
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
22 K. E. Hauschild, J. S. Stover, D. L. Boger and A. Z. Ansari,
Y. Jiang, N. Kalra, B. S. Dwarakanath, U. Langel and
Bioorg. Med. Chem. Lett., 2009, 19, 3779–3782.
M. Ganguli, Mol. Pharmaceutics, 2011, 8, 1729–1741.
23 C. D. Carlson, C. L. Warren, K. E. Hauschild, M. S. Ozers, 25 H. Y. Kuchelmeister, S. Karczewski, A. Gutschmidt,
N. Qadir, D. Bhimsaria, Y. Lee, F. Cerrina and A. Z. Ansari,
Proc. Natl. Acad. Sci. U. S. A., 2010, 107, 4544–4549;
S. Knauer and C. Schmuck, Angew. Chem., Int. Ed., 2013,
52, 14016–14020.
W. C. Tse and D. L. Boger, Acc. Chem. Res., 2004, 37, 26 S. Foillard, M. O. Rasmussen, J. Razkin, D. Boturyn and
61–69. P. Dumy, J. Org. Chem., 2008, 73, 983–991.
24 G. Gasparini, E.-K. Bang, J. Montenegro and S. Matile, 27 C. Kay, O. E. Lorthioir, N. J. Parr, M. Congreve,
Chem. Commun., 2015, 51, 10389–10402; A. Mann,
G. Thakur, V. Shukla, A. K. Singh, R. Khanduri, R. Naik,
S. C. McKeown, J. J. Scicinski and S. V. Ley, Biotechnol.
Bioeng., 2000, 71, 110–118.
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2015