Targeting of polyamidoamine-DNA nanoparticles using the Staudinger ligation: Attachment of an RGD motif either before or after complexation

Article abstract of DOI:10.1016/j.bmc.2008.05.023

Two new methods for the modular synthesis of targeted gene delivery systems are reported. The PEGylated polyamidoamine DMEDA-PEG-DMEDA-(MBA-DMEDA)_n+1-PEG-DMEDA 3 was sequentially modified to contain an integrin-binding peptide ligand via the Staudinger ligation. The conjugation of the ligand was achieved either before particle complexation (precomplexation) or after particle complexation (postcomplexation). Comparison of the two systems showed that postcomplexation strategy led to small and discrete toroidal nanoparticles whilst the precomplexation particles showed loose complexes. The targeted particles showed an increased uptake into cells compared to unmodified complexes however no significant increase in transfection was seen.

Full text of DOI:10.1016/j.bmc.2008.05.023

Bioorganic & Medicinal Chemistry 16 (2008) 6641–6650
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Targeting of polyamidoamine–DNA nanoparticles using the Staudinger
ligation: Attachment of an RGD motif either before or after complexation
Susan M. Parkhouse a, Martin C. Garnett b, Weng C. Chan a,
*
a School of Pharmacy, Centre for Biomolecular Sciences, University of Nottingham, University Park, Nottingham NG7 2RD, UK
b School of Pharmacy, Boots Science Building, University of Nottingham, University Park, Nottingham NG7 2RD, UK
a r t i c l e i n f o
a b s t r a c t
Article history:
Two new methods for the modular synthesis of targeted gene delivery systems are reported. The PEGy-
lated polyamidoamine DMEDA-PEG-DMEDA-(MBA-DMEDA)n+1-PEG-DMEDA 3 was sequentially modi-
fied to contain an integrin-binding peptide ligand via the Staudinger ligation. The conjugation of the
ligand was achieved either before particle complexation (precomplexation) or after particle complexation
(postcomplexation). Comparison of the two systems showed that postcomplexation strategy led to small
and discrete toroidal nanoparticles whilst the precomplexation particles showed loose complexes. The
targeted particles showed an increased uptake into cells compared to unmodified complexes however
no significant increase in transfection was seen.
Received 3 January 2008
Revised 30 April 2008
Accepted 7 May 2008
Available online 10 May 2008
Keywords:
Target gene delivery
Polyamidoamine
RGD peptides
Ó 2008 Elsevier Ltd. All rights reserved.
Staudinger ligation
1. Introduction
ies. The 10-residue sequence ensures a native spacer chain be-
tween the integrin-binding ‘RGD’ motif and the chemical ligation
Polymer based non-viral gene delivery systems have been stud-
ied as a potential way of delivering genes to faulty cells. Several
polymer based delivery systems have been studied including
site. This sequence has previously been used as a competitive
inhibitor of fibronectin7b and it binds well with integrin receptors.
Amongst the numerous methodologies for ligand attachment,
we considered that the ligation based on the Staudinger reaction9
is one of the most robust and chemospecific. The ‘Staudinger liga-
tion’ is a reaction between an appropriately modified phosphine
and an azide, in which an amidophosphonium intermediate is
hydrolyzed in water to form an amide bond. It was independently
developed by Bertozzi10 and Raines11 and has many applications in
chemical biology.12 A particular advantage of using the Staudinger
ligation is that the ligand can be readily attached either before
assembly of the complexes (precomplexation) or to the assembled
nanoparticle (postcomplexation). This will enable us to explore
which of these strategies is most advantageous in terms of both
the chemistry and the assembly of complexes to produce the most
effective delivery vehicles.
In our strategy, the linear block polymer, a PEGylated polyami-
doamine has been N-acylated with 2-azidoacetic acid, which effec-
tively acts as a masked glycine residue. The 10-residue native
peptide chain can then be ligated to the modified linear polymer.
Postcomplexation ligation of the peptide should guarantee that
the targeting ligand is on the outside and minimize the disruption
of the particles. Precomplexation ensures that the ligation is easy
to monitor and that the products are well defined. However, the
latter route has the adverse possibilities for the ligand to be
sequestered inside the complex and thus unavailable for binding,
and also for the ligand to affect the complexation between the
poly-L
-lysine,1 polyethyleneimines,2 polyamidoamine dendrimers3
and linear polyamidoamines,4 all of which have had some success
as suitable delivery vehicles. The efficiency of non-viral systems
could be markedly improved by attachment of ligands recognising
specific receptors on cell surfaces. Ligand targeting can enhance
both specificity and uptake. Some examples of this include poly-
saccharides and peptides.1a,2b,4c
Poly(ethylene glycol)-functionalised (PEGylated) polyamido-
amine:N,N0-methylenebisacrylamide-alt-N,N0-dimethylethylenedi-
amine has been developed for gene delivery5,6 and by targeting this
construct it should be possible to increase the efficiency of uptake
and therefore expression of the gene. There has been extensive
success using RGD-based ligand as targeted delivery via the inte-
grin receptors,7 both for drugs and other gene delivery systems.8
The ‘RGD’ motif is known to bind to several members of the inte-
grin receptor family. Binding of the ligand then triggers receptor
mediated endocytosis. The ‘RGD’–integrin interactions are respon-
sible for diverse cellular functions, including signalling, growth,
differentiation, adhesion and angiogenesis.7 The decapeptide
(RGDSPASSKP) containing the ‘RGD’ motif comprising residues
1615–1624 of fibronectin7 was chosen as the ligand for these stud-
* Corresponding author. Tel.: +44 0115 9515080; fax: +44 0115 9513412.
E-mail address: weng.chan@nottingham.ac.uk (W.C. Chan).
0968-0896/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2008.05.023

6642
S. M. Parkhouse et al. / Bioorg. Med. Chem. 16 (2008) 6641–6650
polymer and DNA. Furthermore, this paper describes for the first
time the successful application of Staudinger ligation between a
phosphinothioester peptide and an azido-bearing block polymer–
DNA complex.
P,S-deprotected immediately prior to condensation with the
synthesised decapeptides. Although the overall yield of the phos-
phothiol reagent 6 was low (7%), each of the chemical transforma-
tions were achieved in modest to good yields (40–69%). In the final
step, best yields were obtained when the saponification was car-
ried out using N2-purged reagents, as well as the use of N2-purged
EtOAc–hexane for column chromatography. Thus, (diphenylphos-
phino)methanethiol 6 was coupled to the protected decapeptides
using carbodiimide-mediated esterification. In order to simplify
the work-up procedure, either the polymer-bound or water-solu-
ble carbodiimide reagents were used. Following acidolytic treat-
ment with a 90% TFA mixture, the thioester peptides 7 and 8
were obtained as white solids in good yields (ca. 90%). Unexpect-
edly, HPLC and MS analysis revealed that racemisation, most likely
epimerisation of the C-terminal amino acid residue (Pro) occurred
prior to the formation of the thioester. Fortuitously, the crucial
‘RGD’ motif is located at the N-terminus, some seven residues from
the C-terminal and is therefore unlikely to affect binding to inte-
grin. In light of this observation, it is recommended that in future
experiments an achiral spacer, for example, glycine or b-alanine,
should be installed at the C-terminus.
The RGDSPASSKP-(diphenylphosphanyl)methyl thiolate 7 and
azidoacetyl functionalised polyamidoamine 5 were successfully li-
gated by stirring in phosphate buffer for 24 h to afford the ‘RGD’-
peptide targeted polymer 9 (precomplexation ligation). The overall
strategy is summarised in Scheme 2. It is worth noting that follow-
ing Staudinger ligation, the azidoacetyl moiety is revealed as a Gly
residue. The linear polymer was then purified by ultrafiltration and
the amount of peptide bound was determined by amino acid anal-
ysis (Table 1). This ligation strategy was then repeated for the ‘non-
binding’ RGRPASSKP-(diphenylphosphanyl)methyl thiolate 8 to af-
ford the ‘RGR’-peptidyl polymer 10.
2. Results
Firstly, the readily prepared polyamidoamine (MBA-DMEDA)n 1
and amine terminated DMEDA-PEG-DMEDA 2 were coupled to-
gether to give the PEGylated polyamidoamine DMEDA-PEG-DME-
DA-(MBA-DMEDA)n+1-PEG-DMEDA 3 (Scheme 1). A large excess
(>10-fold) of the amine component DMEDA-PEG-DMEDA 2 was
used to ensure the formation of desired triblock polymer product
and to minimise uncontrolled multiblock polymerisation. Next,
2-azidoacetic acid 4 was prepared in excellent yield (89%), by
reacting bromoacetic acid with sodium azide. The bis-DMEDA
terminated PEGylated polyamidoamine 3 was then reacted with
carbonyldiimidazole to give a bis(1-imidazolyl carboxamido) inter-
mediate, which on exposure to excess 2-azidoacetic acid afforded
the required azide terminated PEGylated polyamidoamine 5. Here,
the conjugation is via a carbamic anhydride bond. The azido-termi-
nated polymer 5 was then blended with varying ratios of (MBA-
DMEDA)n 1 followed by complexation with the plasmid DNA
gWIZLuc in order to form small, discrete particles as described by
Rackstraw et al.5 Following a series of preliminary experiments,
we established that the 5:1 blend was best achieved at a ratio of
1.75:1 and complexed with gWIZLuc at a polyamidoamine repeat-
ing unit:DNA nucleotide ratio of 2:1. This formulation gave small
toroid like structure as seen by TEM (Fig. 1).
The peptide ligands were synthesised using standard Fmoc
solid-phase synthesis on 2-chlorotrityl poly(styrene) resin
(Scheme 2).13 The decapeptide RGDSPASSKP was chosen as the tar-
geting sequence,7 and RGRSPASSKP as the non-binding sequence as
it has no known binding affinity with integrin receptors. The
assembled protected decapeptides were efficiently cleaved from
the polymer support using a mild acidolytic mixture containing
1% TFA, and stored in their protected forms prior to thioesterifica-
tion with (diphenylphosphino)methanethiol 6. The phosphinothiol
reagent 6 was readily prepared in five steps from thioacetic acid
using a refined method recently developed by Raines and co-work-
ers.11c The synthetic strategy (Scheme 2) involved the preparation
of an air stable borane–diphenylphosphine complex, which was
Using similar conditions, but in distinctively modular manner,
postcomplexation (i.e., using pre-formed polymer blend–gWIZLuc
particle) ligation was successfully performed, and the modified
particle products were purified by dialysis. Amino acid analysis
(Table 1) confirmed the presence of the peptide ligand on both
the precomplexation and the postcomplexation particles with 9%
and 46% substitution for the ‘RGD’-peptidyl products.
TEM (Fig. 1) and DLS (Table 1) studies were carried out on the
precomplexation 9 and 10 formulated gWIZLuc-complexed parti-
cles, and the postcomplexation 11 and 12 particles in order to as-
H
N
H
N
O
HO
OH
O
O
x
O
O
(i) Im2CO, CHCl3
4-MeOPhOH (0.004 eq.), H2O,
MeNH(CH2)2NHMe (0.97 eq.)
(ii) MeNH(CH2)2NHMe (10 eq.)
O
O
O
Me
N
H
N
H
N
Me
N
O
NHMe
O
*
O
N
N
+
N
H
N
H
MeHN
Me
Me
n
x
O
O
O
1
2
H2O, 24 h
O
O
O
Me
N
H
N
Me
N
H
N
O
NHMe
O
*
O
N
O
N
N
N
H
N
H
Me
x
Me
n
Me
O
O
O
3
(i) Im2CO, CH2Cl2, 30 min
(ii) N3CH2CO2H 4, 72 h
O
O
O
H
N-
Me
Me
H
Me
N+
O
N
N
N
N
N
O
O
*
O
N
O
N
N
N
N
N
Me
x
Me
n
Me
H
H
O
O
O
O
O
5
Scheme 1. Synthesis of the 2-azidoacetyl-modified linear block polymer 5.

S. M. Parkhouse et al. / Bioorg. Med. Chem. 16 (2008) 6641–6650
6643
Figure 1. TEM image of ‘naked’ and ‘RGD’-ligated particles. (Top left) Azido-modified PEGylated polyamidoamine 5:(MBA-DMEDA)n 1 (1.75:1) polymer blend/gWIZLuc (2:1)
complex. Magnification: 100,000; scale bar: 200 nm. The arrows point to individual particles. (Top right) Precomplexation-ligation derived particles following ultrafiltration.
Magnification 40,000ꢀ; scale bar 1000 nm. Diameter 55 nm. Small arrows indicate particles, large arrows indicate loose complex. (Bottom) Postcomplexation ligated particles
with purification by dialysis. Magnification 100,000ꢀ; scale bar 200 nm. Diameter 53 nm. Arrows showed discrete spherical particles.
sess their physicochemical characteristics. In all cases, small and
discrete particles were produced. For each particle the Relative
Standard Deviation is given as a measure of peak polydispersity,
where a figure of <20% is considered as monodisperse (i.e., equiv-
alent to a polydispersity index of <0.1).
levels of (MBA-DMEDA)n complexes or PEGylated (MBA-DMEDA)
5 complexes.
3. Discussions
The PEGylated polyamidoamine DMEDA-PEG-DMEDA-(MBA-
Binding and uptake studies were carried out for the precom-
plexation 8 and 9 formulated particles, and postcomplexation li-
gated 10 and 11 particles using 3T3 and A549 cells (Fig. 2).
Fluorescein-labelled RGDPASSKP was used as a binding standard.
YOYO-labelled DNA was used to form and visualise all of the
particles used in the binding/uptake studies. Mono-intercalant
dyes cannot be used in these assays, as the polymer–DNA bind-
ing excludes these dyes from the complex, however, the binding
of bis-intercalant dyes such as YOYO is sufficiently strong to
maintain a high level of labelling. The initial binding and uptake
was low, however after 4 h the fluorescence was high indicating
that significant levels of uptake had taken place in both cell
lines. The difference in fluorescence between 10 min and 4 h
was greater in 3T3 than in A549 cells, but no significant differ-
ences between the ‘RGD’ targeted and ‘RGR’ non-targeted com-
plexes were seen. There was also no significant difference
between the precomplexed ligated and the postcomplexed li-
gated systems.
Finally, the complexes were transfected into 3T3 and A549 cells
(Fig. 3). LipofectAMINETM was used as a positive control, and (MBA-
DMEDA)n–gWIZLuc (5:1) and 5:1 blend–gWIZLuc (2:1) were used
as standards. The overall levels of transfection were low and the
3T3 cells which should have a higher receptor density than the
A549 cells showed lower transfection levels than expected. The
transfection studies showed that targeting with an integrin-bind-
ing peptide ligand did not restore or improve on the transfection
DMEDA)n+1-PEG-DMEDA 3 was synthesised using methods previ-
ously described.5,6 This was then blended with (MBA-DMEDA)n
1
followed by complexation with the plasmid gWIZLuc to produce
well formed discrete nanoparticles, with average size of 75 nm
determined by TEM and 100–200 nm by DLS. The PEGylated parti-
cles were generally round in shape with some toroidal characteris-
tics (Fig. 1). The original publication by Rackstraw et al.5
demonstrates the robustness of this methodology. Unless the for-
mulation is correct, compact toroids are not obtained, and the only
possible formulation under the optimum conditions is one in
which the PEG chains are arranged on the outside of the complex
where DNA is tightly bound at the core.
An azide group was required to perform the Staudinger liga-
tion between the ligand and the PEG modules. The azidoacetyl
moiety was introduced to the terminal amines of 3 through a car-
bamic anhydride, which is envisaged as a biologically acid labile
linker to aid the release of the particle, and in the further devel-
opment of the nanoparticle. Carbamic anhydrides have been
shown to be stable in a wide range of solvents14 and have been
used as stable intermediates for the synthesis of leukotriene
antagonists15 and X-ray contrast agents.16 No change was seen
in the size or the morphology of the particles when 2-azidoacetic
acid was attached to the N-termini of DMEDA-PEG-DMEDA-
(MBA-DMEDA)n+1-PEG-DMEDA 3 and the complexes were formed
(data not shown).

6644
S. M. Parkhouse et al. / Bioorg. Med. Chem. 16 (2008) 6641–6650
(i) (CH2O)n, 80 °C;
(ii) PBr3 (57%)
Br
SH
S
O
O
2-chlorotrityl
poly(styrene)
NaH, Ph2PH•BH3 (44%)
BH3
Clt
Fmoc-Ile
O
PPh2
S
(i) 20% v/v piperidine/DMF
(ii) Fmoc solid-phase peptide synthesis
(iii) Boc2O/DMF
O
(iv) TFA—Et3SiH—CH2Cl2 (1:1:98), 1 h
DABCO, PhMe, 40 °C
(69%)
2 M NaOH(aq)
,
MeOH (40%)
PPh2
PPh2
HS
S
Boc-Arg(Pbf)-Gly-Asp(OtBu)-Ser(tBu)-Pro-Ala-Ser(tBu)-Ser(tBu)-Lys(Boc)-Pro-OH
+
6
O
HOBt, DMAP,
N
C
N
TFA—H2O—Et3SiH (45:4:1), 4 h
PPh2
H-Arg-Gly-Asp-Ser-Pro-Ala-Ser-Ser-Lys-Pro
S
7
(N3CH2COOCO)2-[DMEDA-PEG-DMEDA-(MBA-DMEDA)n+1-PEG-DMEDA] 5,
PBS, pH 7
MBA DMEDA
PEG
n+1
O
O
O
O
O
O
NH
NH
9
Scheme 2. Synthesis of the ‘RGD’-containing decapeptidyl (diphenylphosphanyl)methyl thiolate 7 and the Staudinger ligation product peptidylglycyl PEGylated polyami-
doamine block polymer 9.
Table 1
Dynamic light scattering, TEM and amino acid substitution for postcomplexation and precomplexation ligated DNA-complexed particles
DLS
Polydispersity (relative standard deviation)
TEM
Amino acid substitution
Precomplexation ligation
Postcomplexation ligation
RGD (8)
RGR (9)
118 nm
103 nm
13.1
22.6
55 nm
71 nm
9%
0.4%
RGD (10)
RGR (11)
146 nm
94 nm
5.2
6.4
53 nm
n.a.
46%
n.a.
The (diphenylphosphino)methanethiol 6 was synthesised using
the method developed by Raines11 and gave comparable results.
The reagent was coupled to the ‘RGD’ and ‘RGR’ containing pep-
tides so that the targeting ligand could be ligated to the azide ter-
minated polymer. The Staudinger ligation was carried out under
aqueous conditions, for both the precomplexation and the post-
complexation ligated systems. After 24 h the reaction products
were purified to remove any unreacted peptides and by-products.
This was achieved by ultrafiltration for the precomplexation liga-
tion and dialysis for the postcomplexation ligation, due the small
sample volume. Attachment of the peptide was assessed by amino
acid analysis which showed the peptide had successfully ligated to
the polymer in both the precomplexation and the postcomplex-
ation ligation (Table 1). The 9% ligation was good for the precom-
plexation as the azide was in a relatively low concentration and
the efficiency of the reaction is typically considered to be lower
in water compared to a THF–water mix.11b Surprisingly, the liga-
tion was nearly five times higher, 46%, when assembled complexes
were used. We speculate that this is due to the azide functionality
being more accessible on the surface of the complexes allowing the
ligation to occur more efficiently. Precomplexation ligation using
the ‘RGR’-containing thioester peptide 8 was of unexpectedly low
efficiency (typically 0.4–2%), which is likely to be a result of reac-
tion variability and no attempts were made to optimise reaction
conditions. However, it is worth noting that, compared to the
‘RGD’-containing peptide 7, the peptide 8 is anticipated to be
highly cationic which may indeed negatively affect its ligation effi-
ciency with the cationic polyamidoamine block polymer.
DLS showed that small particles were formed, in which the
postcomplexation ligation particles were observed to be slightly
larger than the precomplexation particles. This may be due the
addition of the peptide on the surface affecting the hydration of

S. M. Parkhouse et al. / Bioorg. Med. Chem. 16 (2008) 6641–6650
6645
25
20
15
10
5
3T3 (10 min)
3T3 (4 h)
A549 (10 min)
A549 (4 h)
0
peptide
unmodified
RGD precomplexation RGD postcomplexation RGR postcomplexation
Figure 2. Uptake studies using A549 and 3T3 cells. The fluorescence shift is used as an indication of the amount of complex taken up by the cells.
100000
A549
3T3
10000
1000
100
10
1
0.1
Figure 3. Transfection of A549 cells (n = 6) and NIH 3T3 cells (n = 6) using ligated block-polyamidoamine–gWIZLuc plasmid (2:1) complexes.
the particles or the thickness of the sterically stabilised PEG layer.
However, there was no significant difference in size when the par-
ticles were measured by TEM (Table 1). TEM showed that the pre-
complexation ligated particles appeared to display some loose
complexation and were not toroidal or spherical as seen with the
‘naked’ PEGylated polyamidoamine–plasmid complex (Fig. 1). This
implied that the presence of the peptide was disrupting the con-
densation between the DNA and the MBA-DMEDA backbone, lead-
ing to loosely formed nanoparticles. Adjusting the formulation may
lead to better condensed and well formed particles. The postcom-
plexation ligated particles formed small particles with toroidal
characteristics similar to the PEGylated nanoparticles with a simi-
lar diameter determined by DLS.
For the cell uptake studies, fluorescein-labelled RGDSPASSKP, in
which the 5-carboxyfluorescein was installed at the Lys-e-amino
group, was synthesised and used to gain an idea of how many
receptors were present on the cell. The integrin receptor expression
was found to be low. However, the level of fluorescence is in the ex-
pected range for the density of receptors reported for the cell type
used.17 The precomplexation and the postcomplexation particles,
prepared using YOYO-labelled gWIZLuc, were incubated with the
cells for 10 min and 4 h in order to gauge the binding and uptake
properties of the complexes, respectively (Fig. 2). The unmodified
complexes, which have no targeting moiety, gave an indication of
non-specific uptake. With the 3T3 cells, both the RGD precomplex-
ation and postcomplexation ligated particles showed greater up-

6646
S. M. Parkhouse et al. / Bioorg. Med. Chem. 16 (2008) 6641–6650
take than the ‘naked’ unmodified or the RGR-ligated complexes
after 10 min. This implied that receptor binding is occurring.
In contrast, after 4 h incubation there was a substantial increase
in the binding and uptake of the complexes compared to the short
incubation time. Only the precomplexation ligated particles, in the
A549 cell line showed a higher uptake than the unmodified com-
plex. The 3T3 cell uptake and the postcomplexation ligated A549
cell uptake was less than unmodified complex uptake. This implied
that, following prolonged exposure, non-specific uptake is likely
the main method of internalisation of complexes.
Flash column chromatography was carried out using silica gel
60 (40–63 lm) from Merck. Analytical TLC was performed using
precoated silica gel 60 (F254) aluminium sheets from Merck and
examined at 254 nm or by permanganate stain. HPLC was carried
out using Waters 510 pumps, Waters 484 detector and Kromasil
C8 analytical column (150 ꢀ 4.6 mm, 5 lm) at a flow rate of
1.0 ml minꢁ1 and the effluent was monitored at 220 nm. Gradient
elution was from 10% to 50% B in 23 min, and the solvents used
were solvent A (0.06% v/v TFA in Milli-Q water) and solvent B
(0.06% v/v TFA in MeCN–Milli-Q water, 9:1 v/v).
Transfections of the polymer blend–gWIZLuc complexes into
A549 cells and 3T3 cells showed no significant change in the
transfection levels compared to the parent azido PEGylated pol-
yamidoamine 5 and did not restore the transfection levels seen by
(MBA-DMEDA)n. The small increase seen in the 3T3 cells is due to
increased non-specific interaction caused by the peptide on the sur-
face of the particles.18 The observed good uptake but poor transfec-
tion could be the result of the particles entering a recycling pathway
and not the lysosomal pathway, hence the plasmid is not delivered
effectively to the cell interior. Good uptake but poor transfection
has been reported previously by folate targeted DNA particles and
was attributed to uptake by a different pathway, resulting in the
plasmid not being delivered effectively to the cell interior.19 An-
other explanation is that the azido PEGylated polyamidoamine 5
particles show higher transfection than has been seen previously,
and is comparable with the (MBA-DMEDA)n complexes.5,6 These
particles have been optimised to form neat, toroidal and spherical
structures. The well formed particles and the addition of the peptide
ligand may have inhibited the release of the complexed plasmid. A
similar result has been reported by Clements et al.,20 where the use
of RGD ligand did not increase the transfection levels of their com-
plex, They suggested that this may be due to inhibited binding,
however our results suggest poor release of the plasmid DNA, as
we have shown both good particle formation and binding of the
targeted complexes to A549 and 3T3 cells.
Peptide synthesis was carried out in a 1 ꢀ 10 cm column (Kine-
sis) using a LKB NovaSyn Gem manual peptide synthesizer at a typ-
ical flow rate of 2.2 ml minꢁ1 with post-column UV monitoring at
290 nm using a LKB Biochrom Ultraspec 4050 spectrometer. Acyl-
ation reactions were performed at ambient temperature and the
mixtures were intermittently stirred.
Luminescence was measured on a Turner TD-20E luminometer.
Protein concentration was measured at 595 nm using a Beckman
DU-640 UV spectrophotometer and analysed using Kineticalc ver-
sion 2.7. Dynamic light scattering was measured using either a
Malvern 4700 photo correlation spectrometer or Viscotek DLS with
Omnisize analytical software. Flow cytofluorimetry was carried out
using a Beckman coulter cytometer and analysed using EXPO 3.2
software.
5.2. Chemicals
Chemicals, biochemical reagents and solvents were purchased
from Sigma–Aldrich (Gillingham, UK) and Fischer Scientific UK Ltd
(Loughborough, UK). Fmoc amino acids, resin support and coupling
reagents were purchased from Novabiochem Merck Biosciences Ltd
(Nottingham, UK). Peptide synthesis grade DMF was purchased
from Rathburn Chemicals Ltd (Walkerburn, Scotland). The plasmid
gWIZLuc, containing the firefly luciferase gene was purchased from
Aldevron (South Fargo, USA) as a 1 mg mlꢁ1 solution.
5.3. Synthetic procedures
4. Conclusions
5.3.1. Polyamidoamine:(N,N0-dimethylethylenediamine-alt-
N,N0-methylenebisacrylamide)-N-propionamidomethyl-
acrylamide, (MBA-DMEDA)n (1)
The azido-terminated PEGylated polyamidoamine 5 polymer
and the 5:1 blend–DNA nanoparticles have been successfully con-
jugated with a peptide ligand via the Staudinger reaction. Formu-
lation has led to well formed, discrete nanoparticles with a small
size. The postcomplexation Staudinger ligation, demonstrated for
the first time in such a complex chemical environment, was supe-
rior to the precomplexation ligation both for efficiency of ligand
conjugation and in terms of ensuring surface localisation of ligand.
While a low level of binding of complexes to cells and uptake into
cells were seen, we could demonstrate neither specificity nor in-
creased transfection activity of the RGD decorated complexes. De-
spite these disappointing results with the RDG ligand, this modular
method of ligand attachment could be used to target other recep-
tors in order to improve the transfection efficiency of non-viral
gene delivery systems.
To a suspension of methylenebisacrylamide (3.10 g, 20.1 mmol)
and 4-methoxyphenol (0.010 g, 0.08 mmol) in distilled H2O
(10 ml) at 0 °C, N,N0-dimethylethylenediamine (2.08 ml, 19.6 mmol)
was added with vigorous stirring under N2 atmosphere. The mixture
was brought up to 27 °C and stirred for 30 min before being left to re-
act in the dark for 2 days. The viscous solution was diluted with
water and acidified to pH 2 with 1.0 M HCl(aq). The solutions were
stored at 4 °C before being purified by ultrafiltration and lyophilised
to afford a white plastic like solid (4.5 g). GPC: number average
molecular weight, Mn 5854, weight average molecular weight, MW
32,037, most frequent mass, MP 22,027, polydispersity index, PDI
5.47; dH (400 MHz, D2O) 2.73, 2.87 (s, NCH3), 3.44, 3.65, 4.50 (s,
NCH2N); dC (100 MHz, D2O) 29.09, 40.51, 44.22, 50.22, 52.70, 171.56.
5. Experimental
5.3.2. Poly(ethylene glycol)bis(N-methyl-2-(methylamino)ethyl
carbamate), DMEDA-PEG-DMEDA (2)
5.1. Instrumentation
Poly(ethylene glycol) 2000 (3.5 g, 1.75 mmol) and carbonyldi-
imidazole (1.16 g, 7.1 mmol) was dissolved in CHCl3 (20 ml) and
stirred for 20 min and N,N0-dimethylethylenediamine (1.86 ml,
17.5 mmol) was added. The solution was stirred overnight before
being concentrated and triturated with ether to produce a white
precipitate. This was collected by filtration and dried under vac-
uum to give the title product (2.75 g, 70%) as a white powder. dH
(400 MHz, D2O) 2.11 (s, NCH3), 3.59 (s, OCH2CH2O), 4.70 (s,
NCH2CH2).
NMR spectra were recorded on a Bruker Ultrashield 400 spec-
trometer. The 13C spectra were proton decoupled. Chemical shifts
are reported in parts per million using a residual protic solvent
as an internal standard and coupling constants (J) are in Hz. Mass
spectra were recorded on a Micromass LCT. UV spectra were re-
corded using WPA lightwave UV machine and glass cuvettes with
a path length of 1.0 cm.

S. M. Parkhouse et al. / Bioorg. Med. Chem. 16 (2008) 6641–6650
6647
5.3.3. Poly(ethylene glycol)bis(N-methyl-2-(methylamino)ethyl
carbamate)-block-polyamidoamine:(N,N0-dimethylethylenedi-
amine-alt-N,N0-methylenebisacrylamide)-N-propionamido-
methyl-acrylamide-block-poly(ethylene glycol)bis(N-methyl-2-
(methylamino)ethyl carbamate), DMEDA-PEG-DMEDA-(MBA-
DMEDA)n+1-PEG-DMEDA (3)
protected amino acids, and activated using TBTU/HOBt/DIEA
(1:1:2 equiv) in DMF; the activated amino acids were used in five-
fold excess and the coupling reactions were typically allowed to
proceed for 2–3 h. Repetitive Fmoc-deprotection was effected by
20% v/v piperidine in DMF for 10–15 min. The N-terminal capping
was achieved using di-tert-butyl dicarbonate (0.655 g, 3.0 mmol)
in DMF for 2 h. The protected peptidyl resin was collected and
dried in vacuo, followed by treatment with TFA/Et3SiH/CH2Cl2
(1:1:98 v/v, 10 ml) for 1 h at ambient temperature. The mixture
was filtered in a solution of pyridine (150 ll) in MeOH (7.5 ml).
The filtrate was evaporated to dryness in vacuo, triturated with
water and the residual material was collected to give the protected
peptide (0.236 g, 47%) as a white solid. A portion of the protected
peptide (5 mg) was treated with TFA/H2O/Et3SiH (45:4:1 v/v,
2.5 ml) for 2 h, followed by evaporation to dryness and trituration
with diethyl ether to afford H-Arg-Gly-Asp-Ser-Pro-Ala-Ser-Ser-
Lys-Pro-OH (2 mg) as a white solid. HPLC: tR 3.66 min; m/z (ES+,
The (MBA-DMEDA)n 1 (0.95 g, 0.029 mmol, MW 32,037) was dis-
solved in distilled H2O and DMEDA-PEG-DMEDA
0.35 mmol) was added. The solution was stirred for 24 h at room
temperature before the pH was adjusted to pH 2 using 1 M HCl(aq)
2 (0.77 g,
.
The triblock polymer was then purified by ultrafiltration and lyo-
philised to yield the title polymer (0.797 g). GPC: Mn 10,707, MW
39,849, MP 28,436, PDI 3.72; dH (400 MHz, D2O) 2.73, 2.87, 3.44,
3.58, 3.65, 4.50; dC (100 MHz, D2O) 29.09, 40.51, 44.22, 50.22,
52.70, 69.54, 171.56.
5.3.4. 2-Azidoacetic acid (4)
HRMS) found MH+ 1001.5825 Calcd for C40H69 N14O16 1001.5011.
þ
Sodium azide (0.98 g, 15 mmol) was dissolved in water (10 ml)
and cooled on an ice bath. Bromoacetic acid (1.04 g, 7.5 mmol) was
added to the mixture and was stirred at 5 °C for 2 h before being
brought up to room temperature and allowed to react for 24 h.
The mixture was acidified to pH 5 with 1 M HCl(aq), extracted into
diethyl ether (3ꢀ 30 ml). The ethereal extract was dried and evap-
orated to dryness to give 2-azidoacetic acid (0.730 g, 89%) as a pale
yellow oil. mmax (cmꢁ1) 1662s, 2110w; dH (400 MHz, CDCl3) 2.51 (s,
CH2); dC 49.99, 174.44; m/z (ES+, HRMS) found MH+ 102.1008,
5.3.8. Boc-Arg(Pbf)-Gly-Arg(Pmc)-Ser(tBu)-Pro-Ala-Ser(tBu)-
Ser(tBu)-Lys(Boc)-Pro-OH
Fmoc-Pro-oxychlorotrityl poly(styrene) (0.282 g, 0.20 mmol,
0.706 mmol gꢁ1) was used for the construction of the protected
‘RGR’-decapeptide by SPPS methodology13 outlined above. The
protected peptidyl resin was collected, dried in vacuo and treated
with TFA/Et3SiH/CH2Cl2 (1:1:98 v/v, 10 ml) for 1 h at ambient tem-
perature. The mixture was filtered in a solution of pyridine (150 ll)
in MeOH (7.5 ml). The filtrate was evaporated to dryness in vacuo,
triturated with water and the residual material was collected to
give the title protected peptide (0.270 g, 70%) as a white solid. A
portion of the protected peptide (5 mg) was treated with TFA/
H2O/Et3SiH (45:4:1 v/v, 2.5 ml) for 2 h, evaporated to dryness
and triturated with diethyl ether to afford H-Arg-Gly-Arg-Ser-
Pro-Ala-Ser-Ser-Lys-Pro-OH (2.4 mg) as a white solid. HPLC: tR
þ
Calcd for C2H4N3O2 102.0298.
5.3.5. N-2-Azidoacetyl-poly(ethylene glycol)bis(N-methyl-2-
(methylamino)ethyl carbamate)-block-polyamidoamine: (N,N0-
dimethylethylenediamine-alt-N,N0-methylenebisacrylamide)-N-
propionamidomethyl-acrylamide-block-N-2-azidoacetyl-poly-
(ethylene glycol)bis(N-methyl-2-(methylamino)ethyl carbamate)
(5)
DMEDA-PEG-DMEDA-(MBA-DMEDA)n+1-PEG-DMEDA
3
3.74 min; m/z (ES+, HRMS) found MH+ 1042.4474 Calcd for
þ
(100 mg, 0.0025 mmol) was dissolved in CH2Cl2 and carbonyldiim-
idazole (0.5 mg, 0.0030 mol) was added. The solution was stirred
for 30 min, after which azidoacetic acid 4 (2.5 mg, 0.025 mmol)
was then added and the resultant solution was stirred for a further
72 h. The solution was concentrated, diluted with water (10 ml),
washed with chloroform (3ꢀ 20 ml). The aqueous extract was then
evaporated to dryness, the residual material triturated with diethyl
ether, and finally redissolved in distilled water (20 ml). The solu-
tion was lyophilised to give the title compound as a white solid
(82 mg). mmax (cmꢁ1) 1662s, 2110w.
C
42H76 N17O14 1042.5752.
5.3.9. H-Arg-Gly-Asp-Ser-Pro-Ala-Ser-Ser-Lys(fluorescein-5-
carbonyl)-Pro-OH
Fmoc-Pro-oxychlorotrityl poly(styrene) (0.144 g, 0.10 mmol,
0.686 mmol gꢁ1) was used for the construction of above fluores-
cein-labelled ‘RGD’-peptide. The Lys residue was installed using
the pseudo-orthogonally protected Fmoc-Lys(ivDde)-OH.21 Follow-
ing assembly and N-capping, the N-ivDde group was removed
using 3–5% v/v NH2NH2ꢂH2O in DMF for 10 min. The peptidyl resin
was then treated with 5-carboxyfluorescein (94 mg, 0.25 mmol)–
7-aza-1-hydroxybentriazole, HOAt22 (34 mg, 0.25 mmol)–N,N0-
diisopropylcarbodiimide (0.046 ml, 0.3 mmol) in DMF (2 ml) and
stirred gently overnight. The resin product was then washed with
20% v/v piperidine in DMF (2 ml minꢁ1) for 10–15 min, washed
with DMF, collected and dried in vacuo. A sample of the peptidyl
resin (103 mg) was treated with TFA/H2O/Et3SiH (45:4:1 v/v,
10 ml) for 2 h, followed by evaporation to dryness and trituration
with diethyl ether to afford the titled fluorescein-labelled ‘RGD’-
peptide (31 mg) as a bright yellow solid. HPLC: tR 10.83 min; m/z
5.3.6. N-9-Fluorenylmethoxycarbonyl-prolyl-oxychlorotrityl
poly(styrene)
The 2-chlorotrityl chloride poly(styrene) resin (0.429 g,
0.60 mmol, 1.4 mmol gꢁ1, 100–200 mesh, 1% DVB)13 and Fmoc-
Pro-OH (0.202 g, 0.6 mmol) were placed in a round bottom flask
and dried in vacuo for 1 h. The solvent CH2Cl2 (10 ml) was added
followed by N,N-diisopropylethylamine (154 ll, 0.90 mmol, DIEA),
and the resultant mixture was stirred for 2 h. MeOH (1.0 ml) was
added and the mixture stirred for a further 10 min. The resin was
collected, washed with DMF (10ꢀ 1 ml), CH2Cl2 (10ꢀ 1 ml) and
hexane (10ꢀ 1 ml), dried in vacuo to give the resin product
(0.606 g) and stored at 4 °C. The amino acid loading, determined
(ES+, HRMS) found MH+ 1359.4891 Calcd for C61H79N14O22
þ
1359.5488.
by Fmoc substitution level13b was found to be 0.78 mmol gꢁ1
.
5.3.10. S-Bromomethyl ethanethiolate
A mixture of thioacetic acid (4.00 g, 58 mmol) and paraformal-
dehyde (1.78 g) heated at 80 °C overnight. The mixture was then fil-
tered by gravity and the resultant yellow oil was cooled in an ice
bath under nitrogen. Phosphorus tribromide (10.70 g, 39.5 mmol)
was added slowly keeping the temperature <5 °C. The solution
was stirred at 0 °C for 10 min before being warmed to room temper-
ature. The mixture was then poured into ice water (10 ml) and ex-
5.3.7. Boc-Arg(Pbf)-Gly-Asp(OtBu)-Ser(tBu)-Pro-Ala-Ser(tBu)-
Ser(tBu)-Lys(Boc)-Pro-OH
Fmoc-Pro-oxychlorotrityl poly(styrene) (0.385 g, 0.30 mmol,
0.78 mmol gꢁ1) was used for the construction of the protected dec-
apeptide by standard solid-phase peptide synthesis (SPPS) meth-
odology.13 The amino acid residues were installed using N-Fmoc-

6648
S. M. Parkhouse et al. / Bioorg. Med. Chem. 16 (2008) 6641–6650
tracted into diethyl ether (3ꢀ 20 ml). The organic extract was
washed with water (2ꢀ 50 ml), dried, concentrated and purified
by column chromatography (silica, ether) to give pure S-bromo-
methyl ethanethiolate (5.57 g, 57%) as a yellow oil. The product
was stored in a nitrogen atmosphere below 4 °C. dH (400 MHz,
CDCl3) 2.42 (3H, s, CH3) 4.75 (2H, s, CH2) TLC: Rf 0.775 (ether).
form was added N-poly(styrene)methyl-N0-cyclohexylcarbodii-
mide (76 mg, 0.13 mmol; 1.7 mmol gꢁ1, 200–400 mesh, 2% DVB),
HOBt (9.6 mg, 0.071 mmol) and 4-(N,N-dimethylamino)pyridine,
DMAP (1.0 mg, 0.008 mmol). The mixture was stirred for 30 min,
after which (diphenylphosphino)methanethiol
6
(45 mg,
0.19 mmol) was added, and the mixture was stirred for a further
48 h. The mixture was filtered, the filtrate evaporated to dryness,
and the residual material triturated with water. The resulting solid
was dissolved in the mixture TFA/H2O/Et3SiH (45:4:1 v/v, 10 ml)
and left for 4 h. The mixture was evaporated to dryness and the
residual material was triturated with diethyl ether to give an
off-white solid. The solid was dissolved in water and lyophilised
to yield H-Arg-Gly-Asp-Ser-Pro-Ala-Ser-Ser-Lys-Pro-SCH2PPh2
(42 mg, 88%). HPLC: tR 16.76, 17.34 min (2:1); m/z (ES+, HRMS)
found MH+ 1215.5190 and 1215.5458 Calcd for C53H80N14O15PS+
1215.5380.
5.3.11. Borane–S-(diphenylphosphanyl)methyl ethanethiolate
complex
NaH (0.754 g, 22.4 mmol, 70% dispersion in oil) was placed in a
two-necked round bottom flask connected to a nitrogen bubbler.
The NaH was washed three times with hexane, dissolved in DMF
(10 ml) and the mixture was cooled to 0 °C. Borane diphenylphos-
phine complex (4.5 g, 22.4 mmol) in DMF (3 ml) was added slowly
keeping the mixture below 5 °C. The S-bromomethyl ethanethiolate
(3.74 g, 22.4 mmol) was then added and the mixture stirred at 0 °C
for 10 min before being allowed to warm to room temperature.
The mixture was stirred for a further 24 h, concentrated, redissolved
in EtOAc (70 ml), filtered and evaporated to dryness. The crude prod-
uct was purified by column chromatography (silica, EtOAc/hexane
1:10), to give required borane–thiolacetic acid ester complex11c
(2.854 g, 44%) as a clear oil. dH (400 MHz, CDCl3, P decoupled)
0.51–1.68 (3H, br m, BH3), 2.28 (3H, s, CH3), 3.74 (2H, d, J = 6 Hz,
CH2), 7.47–7.53 (6H, m, Ph) 7.73–7.74 (4H, m, Ph); dC (100 MHz,
CDCl3) 24.15 (d, J = 35.5 Hz, 30.46, 127.97 (d, J = 55.2 Hz), 129.28
(d, J = 10.2 Hz), 132.19 (d, J = 2.6 Hz), 132.85 (d, J = 9.3 Hz), 193.64;
m/z (ES+, HRMS) found MH+ 289.0509 Calcd for C15H19BOPS+
289.0981.
5.3.15. H-Arg-Gly-Arg-Ser-Pro-Ala-Ser-Ser-Lys-Pro-
(diphenylphosphanyl)methyl thiolate (8)
A
solution of Boc-Arg(Pbf)-Gly-Arg(Pmc)-Ser(tBu)-Pro-Ala-
Ser(tBu)-Ser(tBu)-Lys(Boc)-Pro-OH (15 mg, 0.008 mmol) in chlo-
roform (4 ml) was added N-ethyl-N0-(3-dimethylaminopro-
pyl)carbodiimideꢂHCl (4 mg, 0.021 mmol), HOBt (2.25 mg,
0.016 mmol) and DMAP (0.3 mg, 0.0024 mmol), and resultant
mixture was allowed stirred for 30 min. (Diphenylphosphino)-
methanethiol 6 (10 mg, 0.043 mmol) was then added and the
mixture was stirred for 48 h. The solution was evaporated to
dryness and the residual material was triturated with water.
The resulting solid was dissolved in the mixture TFA/H2O/Et3SiH
(45:4:1 v/v, 10 ml) and left for 4 h. The solution was evaporated
to dryness and the residual material was triturated with diethyl
ether to give an off-white solid. The solid was dissolved in water
and lyophilised to afford the peptide thioester H-Arg-Gly-Arg-
Ser-Pro-Ala-Ser-Ser-Lys-Pro-SCH2PPh2 (9.8 mg, 98%). HPLC: tR
14.6 min; m/z (ES+, HRMS) found MH+ 1254.5745 Calcd for
5.3.12. S-(Diphenylphosphanyl)methyl ethanethiolate
The above borane–phosphanylmethyl ethanethiolate complex
(182 mg, 0.633 mmol) was dissolved in toluene (3 ml) and stirred
under N2. DABCO in toluene (3 ml) was added and the mixture
was stirred at 40 °C for 4 h. The solution was then evaporated to
dryness and redissolved in chloroform (20 ml). This solution was
then washed with 1 M HCl (2ꢀ 10 ml) and brine (10 ml), dried,
evaporated to yield (diphenylphosphanyl)methyl thiolacetate11c
(120 mg, 69%) as a clear viscous oil, and was used without further
purification. dH (400 MHz, CDCl3, P decoupled) 2.30 (3H, s, CH3),
3.52 (2H, s, CH2), 7.35–7.36 (6H, m, Ph ) 7.42–7.47 (4H, m, Ph);
dC (100 MHz, CDCl3) 26.38 (d, J = 23.2 Hz), 30.85, 129.33 (d,
J = 7.2 Hz), 129.69, 133.28 (d, J = 18.9 Hz), 137.33 (d, J = 13.7 Hz),
195.22; TLC: Rf 0.15 (EtOAc/hexane 10:1). HPLC: Rf 8.84 min; m/z
(ES+, HRMS) found MH+ 275.0062 Calcd for C15H16OPS+ 275.0654.
C
55H87N17O13PS+ 1256.6122.
5.3.16. N-(H-Arg-Gly-Asp-Ser-Pro-Ala-Ser-Ser-Lys-Pro-Gly)-poly-
(ethylene glycol)bis(N-methyl-2-methylamino)ethyl carbamate)-
block-polyamidoamine:(N,N0-dimethylethylenediamine-alt-N,N0-
methylenebisacrylamide)-N-propionamidomethyl-acrylamide-
block-N-(H-Arg-Gly-Asp-Ser-Pro-Ala-Ser-Ser-Lys-Pro-Gly)-poly-
(ethylene glycol)bis(N-methyl-2-(methylamino)ethyl carbamate)
(9)
The azidoacetyl-modified polymer 5 (39 mg) and H-Arg-Gly-
Asp-Ser-Pro-Ala-Ser-Ser-Lys-Pro-SCH2PPh2 7 (5.0 mg, 0.004 mmol)
were dissolved in phosphate buffer saline (2 ml) and allowed to re-
act overnight. The resulting solution was then purified by ultrafil-
tration at 5000 rpm (3 ꢀ 10 min). The solution was then lyophised
to yield a white plastic like solid (28 mg, 71% w/w). Amino acid
analysis: 9% substitution.
5.3.13. (Diphenylphosphino)methanethiol (6)
The above (diphenylphosphanyl)methyl thiolacetate (119 mg,
0.43 mmol) was dissolved in degassed MeOH (3 ml) and stirred un-
der nitrogen. A solution of aqueous NaOH (5 M in degassed water,
2 ml) was added and the solution was stirred for 2 h before being
concentrated and redissolved in chloroform (20 ml). The organic
solution was washed with a mixture 1 M HCl(aq) and brine (3ꢀ
10 ml), dried over MgSO4 and evaporated to dryness. The crude
product was then rapidly purified by column chromatography (sil-
ica, degassed EtOAc/hexane 3:1) to give the title compound11b
(40 mg, 40%) as a colourless viscous oil. dH (400 MHz, CDCl3, P decou-
pled) 1.40 (1H, t, J = 8.0 Hz, SH), 3.06 (2H, d, J = 7.5 Hz, CH2), 7.36–
7.38 (6H, m, Ph) 7.42–7.45 (4H, m, Ph); dC (100 MHz, CDCl3) 20.86
(d, J = 23.7 Hz), 128.59, 128.65, 129.15, 132.79 (d, J = 18.5 Hz); TLC:
Rf 0.15 (EtOAc/hexane 1:10), Rf 0.66 (EtOAc/hexane 1:3); m/z (ES+,
HRMS) found MH+ 233.0522 Calcd for C13H14PS+ 233.0548.
5.3.17. N-(H-Arg-Gly-Arg-Ser-Pro-Ala-Ser-Ser-Lys-Pro-Gly)-poly-
(ethylene glycol)bis(N-methyl-2-(methylamino)ethyl carbamate)-
block-polyamidoamine:(N,N0-dimethylethylenediamine-alt-N,N0-
methylenebisacrylamide)-N-propionamidomethyl-acrylamide-
block-N-(H-Arg-Gly-Arg-Ser-Pro-Ala-Ser-Ser-Lys-Pro-Gly)-poly-
(ethylene glycol)bis(N-methyl-2-(methylamino)ethyl carbamate)
(10)
The azidoacetyl-modified polymer 5 (25 mg) and H-Arg-Gly-
Arg-Ser-Pro-Ala-Ser-Ser-Lys-Pro-SCH2PPh2 8 (5.0 mg, 0.004 mmol)
were dissolved in phosphate buffer saline (2 ml) and allowed to re-
act overnight. The resulting solution was then purified by ultrafil-
tration at 5000 rpm (3 ꢀ 10 min). The solution was then lyophised
to yield a white plastic like solid (23 mg, 92% w/w). Amino acid
analysis: 0.4% substitution.
5.3.14. H-Arg-Gly-Asp-Ser-Pro-Ala-Ser-Ser-Lys-Pro-
(diphenylphosphanyl)methyl thiolate (7)
To a solution of Boc-Arg(Pbf)-Gly-Asp(tBu)-Ser(tBu)-Pro-Ala-
Ser(tBu)-Ser(tBu)-Lys(Boc)-Pro-OH (65 mg, 0.039 mmol) in chloro-

S. M. Parkhouse et al. / Bioorg. Med. Chem. 16 (2008) 6641–6650
6649
5.4. Fluorescent tagging of DNA
5.9.2. Postcomplexation RGR-modified particles (12)
The 5:1–DNA complexes were prepared and left for 30 min. The
decapeptidyl thioester H-Arg-Gly-Agr-Ser-Pro-Ala-Ser-Ser-Lys-
Pro-SCH2PPh2 8 (ca. threefold excess) was added to the solution
and allowed to react overnight. The resulting solution was then
purified by ultrafiltration 5000 rpm (3 ꢀ 10 min) or by dialysis
overnight.
gWIZLuc (100 ll, 5 mg mlꢁ1) was mixed with YOYO the solution
(20 ll, 1 mM in DMSO) by vortex and incubated at 4 °C for 24 h.
The labelled DNA was stored at 4 °C for future use.
5.5. Gel permeation chromatography
Tris/NaCl solution was prepared from Tris acetate (12.1 g lꢁ1
,
5.10. Particle size analysis
0.1 M) and sodium chloride (11.7 g lꢁ1, 0.2 M) and filtered through
0.22 lm filter and purged with N2. TSK G3000 and TSK G4000 pw
columns were attached to the pump in series and equilibrated with
Tris/NaCl at 0.6 ml minꢁ1. Solutions of PEG 8600, PEG 23000, PEG
40000 (1 mg mlꢁ1) were prepared using Tris/NaCl buffer, and were
used to generate a linear calibration curve. The polyamidoamine
samples were then analyzed to determine the molecular weight
and polydispersity.
Single aliquots of the polymer was added to gWIZLuc (4 ll) and
1/10 PBS (50 ll) in a UvetteÒ and gently mixed. The size and poly-
dispersity was then determined by dynamic light scattering using a
DLS.
5.10.1. Transmission electron microscopy
Complexes were prepared by adding polymer to gWIZLuc (4 ll)
and 1/10 PBS (500 ll) and incubated for 30 min. A drop of the solu-
tion was placed a copper grid coated with Pioloform resin, left for
1 min, the excess liquid was removed and the process repeated.
After air drying, the grid was inverted in uranyl acetate for
20 min. The grid was washed in 50% ethanol(aq) twice followed
by water. The excess water was removed by blotting and the grids
were air dried. The grids were examined using a JEOL JEM-1010
transmission electron microscope. Images were taken between
20,000 and 120,000ꢀ using a Kodak megaplus digital camera and
analysed using Analysis 3.0 software.
5.6. Polymer:gWIZLuc complex preparation
Unless otherwise stated the complexes where prepared by add-
ing a solution of the polymer to a dilute solution of the DNA in the
required amounts to produce the complexes required at the correct
repeating unit (RU):DNA nucleotide ratio. This was calculated from
the following equation:
Amount of polymer ¼ ½RU MWpolymer=RU MW DNAꢃ ꢀ ratio
ꢀ Amount of DNA
5.11. Biological evaluation
The RU MWDNA is taken as 308 (mean molecular weight of the
nucleotides) and RU MWpolymer for MBA-DMEDA as 315.
5.11.1. Cell lines and routine subculture
A549 cells (ECACC No. 86012804) and NIH 3T3 cells (ECACC No.
93061524) were routinely cultured in DMEM supplemented with
10% heat inactivated foetal calf serum, antibiotics and antimicrobi-
5.7. Gel retardation electrophoresis
Complexes were produced at a range of ratios by adding the re-
quired amount of polymer to gWIZLuc (2 ll) and tris acetate buffer
(TAE, 16 ll). The solution was mixed by vortex and allowed to form
for 30 min before gel loading buffer (3 ll) was added. The complexes
(10 ll) were loaded into 0.8% agarose gel (1.6 g in 250 ml TAE) con-
taining 1% ethidium bromide. The gel was developed at 100 V for
60 min using TAE as the running buffer. The gels were then visual-
ised using a UV transilluminator to visualise the DNA before being
stained with Brilliant blue R250 (0.5 g in methanol/glacial acetic
acid/distilled water; 250:50:50 ml) for 1 h and destained (metha-
nol/glacial acetic acid/distilled water; 1:1:1 v/v) overnight.
als, 3T3 cells were also supplemented with -glutamine, at 37 °C
L
and 5% CO2. The cell lines were subcultured every seven days at
a ratio of 1:20, using trypsin EDTA to remove cells from flask
surface.
5.11.2. Cell fixation
A549 and 3T3 cells were grown to confluence in T75 flasks. The
cells were detached using trypsin/EDTA and ultrafitration at
5000 rpm for 5 min. The cells were washed with PBS and sus-
pended in 4% paraformaldehyde(aq) before being incubated at
37 °C for 10 min followed by 4 °C for 10 min. The cells were ultra-
filtered at 5000 rpm for min and washed twice with PBS and stored
at 4 °C.
5.8. Formulation
Polymer blends were prepared using the equation:
5.11.3. Fluorescence microscopy
3T3 and A549 cells were seeded into a 6-well plate at a density
of 105 cells per well, and grown for 24 h. Complexes were prepared
as previously described, 300 ll of the complex was added to DMEM
(3 ml). The media was removed for the cells, washed with PBS and
then replaced with the complexes (1 ml ꢀ 2 wells). The cells were
incubated for 4 h then examined using a Nikon microscope and a
mercury lamp at 500ꢀ magnification.
Vpol 1 ¼ ½MWpol 1=MWpol 2ꢃ ꢀ ratio ꢀ Vpol 2
where V is volume of polymer required and MW is the RU molecular
weight.
The individual polymers were in 1 mg mlꢁ1 stock solutions, and
a single portion of azido-terminated polymer 5 was aliquotted into
polyamidoamine (MBA-DMEDA)n 1.
5.9. Staudinger ligation to postassembled complexes
5.11.4. Cell binding studies
A549 were grown in a T75 flask cells were fixed using parafor-
maldehyde and resuspended in 5 ml of PBS. 500 ll aliquots were
treated with native 3:1-derived particles, unmodified 5:1-derived
particles, RGD precomplex modified particles, RGD postcomplex
modified particles, RGR postcomplex modified particles and RGD
fluorescein-labelled peptide for 10 min. The cells were ultrafiltered
at 5000 rpm for 5 min in an MSE Centaur centrifuge, washed twice
with PBS and resuspended in 300 ll of PBS. The cells were read in a
5.9.1. Postcomplexation RGD-modified particles (11)
The 5:1–DNA complexes were prepared and left for 30 min. The
decapeptidyl thioester H-Arg-Gly-Asp-Ser-Pro-Ala-Ser-Ser-Lys-
Pro-SCH2PPh2 7 (ca. threefold excess) was added to the solution
and allowed to react overnight. The resulting solution was then
purified by ultrafiltration 5000 rpm (3 ꢀ 10 min) or by dialysis
overnight. Amino acid analysis: 48% substitution.

6650
S. M. Parkhouse et al. / Bioorg. Med. Chem. 16 (2008) 6641–6650
cytofluorometer and analysed using EXPO software. The same pro-
cess was repeated using 3T3 cells.
was added, the solutions were incubated in the dark for 15 min be-
fore being read at 595 nm using a plate reader.
5.11.5. Uptake experiments
Acknowledgments
A549 cells were seeded into 6-well plates at 250,000 cells per
well and incubated at 37 °C for 24 h. The native 3:1-derived parti-
cles, unmodified 5:1-derived particles, RGD precomplex modified
particles, RGD postcomplex modified particles, RGR postcomplex
modified particles and the fluorescein-labelled ‘RGD’-peptide
(1 mg mlꢁ1) were prepared and diluted in 1.0 ml DMEM. The com-
plexes were added to the cells and incubated at 37 °C for 10 min.
The cells were washed twice with PBS before being loosened with
trypsin/EDTA (300 ll). The cells were centrifuged at 5000 rpm for
5 min, washed with PBS, resuspended in 4% paraformaldehyde(aq)
(600 ll) and incubated at 37 °C for 10 min followed by 4 °C for
10 min. The cells were centrifuged at 5000 rpm for 5 min washed
twice with PBS and resuspended in PBS (300 ll). The cells were
read by cytofluorometry and analysed using EXPO software. The
method was repeated incubating the cells for 4 h and using 3T3
cells at both 10 min and 4 h.
We thank the BBSRC Engineering & Biological Systems research
committee for the funding of a studentship (BBS/S/A/2003/10418)
to Susan Parkhouse.
References and notes
1. (a) Gottschalk, S.; Cristiano, R. J.; Smith, L. C.; Woo, S. L. Gene Ther. 1994, 1, 185–
191;; (b) Mislick, K. A.; Baldeschwieler, J. D.; Kayyem, J. F.; Meade, T. J.
Bioconjugate Chem. 1995, 6, 512–515.
2. (a) Diebold, S. S.; Kursa, M.; Wagner, E.; Cotton, M.; Zenke, M. J. Biol. Chem.
1999, 274, 19087–19094; (b) Kim, E.-M.; Jeong, H.-J.; Kim, S.-L.; Sohn, M.-H.;
Nah, J.-W.; Bom, H.-S.; Park, I.-K.; Cho, C.-S. Nucl. Med. Biol. 2006, 33, 529–534.
3. (a) Gao, X.; Huang, L. Biochem. Biophys. Res. Commun. 1991, 179, 280–285; (b)
Tang, M. X.; Redemann, C. T.; Szoka, J. F. C. Bioconjugate Chem. 1996, 7, 703–
714; (c) Kukowska-Latallo, J. F.; Bielinska, A.; Johnson, J.; Spindler, R.; Tomilia,
D. P.; Baker, J. R. Proc. Natl. Acad. Sci. U.S.A. 1996, 93, 4897–4902.
4. (a) Bignotti, F.; Sozzani, P.; Ranucci, E.; Ferruti, P. Macromolecules 1994, 27,
7171–7178; (b) Hill, I. C. R.; Garnett, M. C.; Bignotti, F.; Davis, S. S. Biochim.
Biophys. Acta 1999, 1427, 161–174; (c) Simoes, S.; Slepushkin, V.; Gaspar, R.; De
Lima, M. C. P.; Duzgune, N. Gene Ther. 1998, 5, 955–964.
5.11.6. Transfections
5. Rackstraw, B.; Stolnik, S.; Davis, S. S.; Bignotti, F.; Garnett, M. C. Biochim.
Biophys. Acta 2002, 1576, 269–286.
A549 or 3T3 cells were seeded into 6-well plates at a density of
50,000 cells per well and grown for 24 h to reach 60–70% conflu-
ence. Complexes were prepared as previously described in 300 ll
using 4 lg of gWIZLuc and were added to OptiMEM buffer (3 ml).
Complexes of LipofectAMINETM were prepared using 4 lg gWIZLuc
according to the manufacturer’s protocol. The medium was aspi-
rated from the cells and rinsed with PBS (1.0 ml). The complexes
were added to the wells (1.0 ml ꢀ 3 wells) and incubated for 4 h
at 37 °C. The media was then aspirated off the cells, was replaced
with fresh DMEM (2.0 ml) and the cells were then incubated for
a further 48 h. Concentrated (5ꢀ) cell lysis buffer (Promega) was
diluted with distilled water to give a ‘1ꢀ’ solution. The media
was removed from the cells and the cells washed with PBS
(1.0 ml) before cell culture lysis buffer was added (400 ll) and left
for 1 min. The cell lysate was collected into centrifuge vials and
centrifuged at 13,000 rpm for 5 min using a bench top microcentri-
fuge and the supernatant was collected.
6. Ranucci, E.; Ferruti, P. Macromolecules 1991, 24, 3747–3752.
7. (a) Pierschbacher, M. D.; Ruoslahti, E. Science 1984, 309, 30–33; (b) Yamada, K.
M.; Kennedy, D. W. J. Cell Biol. 1984, 99, 29–36; (c) Main, A. L.; Harvey, T. S.;
Baron, M.; Boyd, J.; Campbell, I. D. Cell 1992, 71, 671–678; (d) Panetti, T.;
McKeown-Longo, P. J. Biol. Chem. 1993, 268, 11492–11495; (e) Romberger, E.
Annu. Rev. Cell Dev. Biol. 1996, 12, 697–715.
8. (a) Nishiya, T.; Sloan, S. Biochem. Biophys. Res. Commun. 1996, 224, 242–245; (b)
Temming, K.; Schiffelers, R. M.; Molema, G.; Kok, R. J. Drug Resist. Updat. 2005,
8, 381–402.
9. Staudinger, H.; Meyer, J. Helv. Chim. Acta 1919, 2, 635–646.
10. (a) Saxon, E.; Bertozzi, C. R. Science 2000, 287, 2007–2010; (b) Saxon, E.;
Armstrong, J. I.; Bertozzi, C. R. Org. Lett. 2000, 2, 2141–2143.
11. (a) Nilsson, B. L.; Kiessling, L. L.; Raines, R. T. Org. Lett. 2000, 2, 1939–1941; (b)
Nilsson, B. L.; Kiessling, L. L.; Raines, R. T. Org. Lett. 2001, 3, 9–12; (c) Soellner,
M. B.; Nilsson, B. L.; Raines, R. T. J. Org. Chem. 2002, 67, 4993–4996.
12. (a) Prescher, J. A.; Bertozzi, C. R. Nat. Chem. Biol. 2005, 1, 13–21; (b) Soellner, M.
B.; Dickson, K. A.; Nilsson, B. L.; Raines, R. T. J. Am. Chem. Soc. 2003, 125, 11790–
11791.
13. (a) Barlos, K.; Chatzi, O.; Gatos, D.; Stavropoulos, G. Int. J. Pept. Protein Res. 1991,
37, 513–520; (b)Fmoc Solid Phase Peptide Synthesis: A Practical Approach; Chan,
W. C., White, P. D., Eds.; Oxford University Press: Oxford, 2000.
14. (a) Schroeder, W. A.; Wilcox, P. E. J. Am. Chem. Soc. 1950, 72, 3814–3815; (b)
Shepard, K. L. Chem. Commun. 1971, 928–929.
15. Yee, Y. K.; Bernstein, P. R.; Adams, E. J.; Brown, F. J.; Cronk, L. A.; Hebbel, K. C.;
Vacek, E. P.; Krell, R. D.; Snyder, D. W. J. Med. Chem. 1990, 33, 2437–2451.
16. Illig, C. R. Eastman Kodak Company (Rochester, NY), United States, Patent
5472683, 1995.
17. Nishiuchi, R.; Murayama, O.; Fujiwara, H.; Gu, J.; Kawakami, T.; Aimoto, S.;
Wada, Y.; Sekiguchi, K. J. Biochem. (Tokyo) 2003, 134, 497–504.
18. Colin, M.; Maurice, M.; Trugnan, G.; Kornprobst, M.; Harbottle, R. P.; Knight, A.;
Cooper, R. G.; Miller, A. D.; Capeau, J.; Coutelle, C.; Brahimi-Horn, M. C. Gene
Ther. 2000, 7, 139–152.
5.11.6.1. Luciferase activity. The luciferase activity was measured
by mixing the cell lysate (20 ll) with Luciferin reagent (100 ll) in a
scintillation vial and measured using a luminometer calibrated
with luciferin reagent (100 ll). Each sample was measured three
times and the average value used in analysis.
5.11.6.2. Bradford assay. A solution of lysis buffer (1:10, 10.0 ml)
was prepared and used to make bovine serum albumin standards
of 0.3, 0.2, 0.1, 0.05, 0.025, 0.0125, 0.00625 and 0.003125 mg mlꢁ1
from a stock solution (1 mg mlꢁ1). Solutions of all the samples
were prepared (3ꢀ 200 ll) using the distilled H2O so that they
were diluted by a factor of 10. Samples and standards (10 ll) were
transferred into a 96-well plate and Bradford assay reagent (200 ll)
19. Dauty, E.; Remy, J.-S.; Zuber, G.; Behr, J.-P. Bioconjugate Chem. 2002, 13, 831–
839.
20. Clements, B. A.; Bai, J.; Kucharski, C.; Farrell, L.-L.; Lavasanifar, A.; Ritchie, B.;
Ghahary, A.; Uludag, H. Biomacromolecules 2006, 7, 1481–1488.
21. Chhabra, S. R.; Hothi, B.; Evans, D. J.; White, P. D.; Bycroft, B. W.; Chan, W. C.
Tetrahedron Lett. 1998, 39, 1603–1606.
22. Carpino, L. A. J. Am. Chem. Soc. 1993, 115, 4397–4398.