A click chemistry route to 2-functionalised PEGylated and cationic β-cyclodextrins: Co-formulation opportunities for siRNA delivery

Article abstract of DOI:10.1039/c2ob25490e

A new approach to the synthesis of amphiphilic β-cyclodextrins has used 'click' chemistry to selectively modify the secondary 2-hydroxyl group. The resulting extended polar groups can be either polycationic or neutral PEGylated groups and these two amphiphile classes are compatible in dual cyclodextrin formulations for delivery of siRNA. When used alone with an siRNA, a cationic cyclodextrin was shown to have good transfection properties in cell culture. Co-formulation with a PEGylated cyclodextrin altered the physicochemical properties of nanoparticles formed with siRNA. Improved particle properties included lower surface charges and reduced tendency to aggregate. However, as expected, the transfection efficiency of the cationic vector was lowered by co-formulation with the PEGylated cyclodextrin, requiring future surface modification of particles with targeting ligands for effective siRNA delivery.

Full text of DOI:10.1039/c2ob25490e

View Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
Organic &
Biomolecular
Chemistry
Cite this: Org. Biomol. Chem., 2012, 10, 4954
www.rsc.org/obc
PAPER
A click chemistry route to 2-functionalised PEGylated and cationic
β-cyclodextrins: co-formulation opportunities for siRNA delivery†
Aoife M. O’Mahony,‡a Julien Ogier,‡b Stephane Desgranges,b John F. Cryan,c Raphael Darcyb
and Caitriona M. O’Driscoll*a
Received 6th March 2012, Accepted 1st May 2012
DOI: 10.1039/c2ob25490e
A new approach to the synthesis of amphiphilic β-cyclodextrins has used ‘click’ chemistry to selectively
modify the secondary 2-hydroxyl group. The resulting extended polar groups can be either polycationic
or neutral PEGylated groups and these two amphiphile classes are compatible in dual cyclodextrin
formulations for delivery of siRNA. When used alone with an siRNA, a cationic cyclodextrin was shown
to have good transfection properties in cell culture. Co-formulation with a PEGylated cyclodextrin altered
the physicochemical properties of nanoparticles formed with siRNA. Improved particle properties
included lower surface charges and reduced tendency to aggregate. However, as expected, the transfection
efficiency of the cationic vector was lowered by co-formulation with the PEGylated cyclodextrin,
requiring future surface modification of particles with targeting ligands for effective siRNA delivery.
cultures, namely lowered tendency to aggregate in formulations
Introduction
and good (though non-specific) adhesion to cell surfaces.
However, there are drawbacks associated with such cationic mole-
cules including cellular toxicity and limited circulation in vivo.
One approach to overcoming these drawbacks is the modification
of formulations with poly(ethylene glycol) (PEG), which has
been widely reported.13–15 The PEG provides a surface which is
both neutral and non-aggregating, with reduced toxicity. In
addition, PEG chains are proven as long linkers for exposition of
surface ligands which can guide delivery to cell receptors.16
Here we have devised a new strategy for synthesis of cationic
and neutral PEGylated CDs which possess similar amphiphili-
city, thus enabling co-assembly and co-formulation to provide a
PEGylated surface to cationic CD : siRNA nanoparticles. In each
series, the polar groups are attached by cuprous-catalysed ‘click’
chemistry. Previously click chemistry has been used on the
primary side of β-CD gene delivery vectors,17–20 with only
limited reports of its use on the secondary side.21 The extension
of any such chemistry to the secondary side with its two sets of
hydroxyl groups presents a challenge in selectivity, as options
are limited for introduction of groups at the 2-positions only.22
We previously grafted short oligo(ethylene oxide) groups at
these positions by base-catalysed reaction with ethylene carbon-
ate.3,23 However we found this reaction of limited use for chains
longer than about two to three ethylene oxide units. This may be
due to steric hindrance or the formation of ester rather than oxo
links. Photochemical addition of thiols to 2-alkenylated CDs has
lent itself well to the introduction of lipid groups at the 2-posi-
tions,6 but is a less convenient option for the attachment of PEG.
Here, while thioalkylation is used to provide the hydrophobic
groups on the primary-OH face as before,3,23 copper(I)-catalysed
Small interfering RNAs (siRNAs) are capable of specific and
efficient gene silencing,1 via the intracellular RNA interference
(RNAi) pathway and are particularly useful for studying gene
function. They also present a promising therapeutic strategy in a
wide variety of diseases. As oligonucleotides however, siRNAs
show poor cell penetration and limited stability in a physiologi-
cal environment,2 therefore considerable research has focused on
development of vectors for their delivery.
Cyclodextrins (CDs), cyclomalto-oligosaccharides, are natu-
rally occurring oligosaccharides which are well established
pharmaceutical exipients. They offer many advantages as cores
for the synthesis of non-viral vectors and are by now well estab-
lished as potential delivery agents for therapeutic
oligonucleotides.3–5 Amphiphilic cationic cyclodextrins have
been particularly promising.3,6–10 The choice of cationic group
and length of amphiphilic chain have both been shown to
influence binding of pDNA, properties of the complexes formed
and transfection efficiency.4,8,11,12
Cationic vectors for siRNA delivery offer several advantages
which usually enhance the overall results for transfection in cell
aSchool of Pharmacy, University College Cork, Cork, Ireland.
E-mail: caitriona.odriscoll@ucc.ie
bCentre for Synthesis and Chemical Biology, School of Chemistry,
University College Dublin, Dublin 4, Ireland
cDepartment of Anatomy and Neuroscience, University College Cork,
Cork, Ireland
†Electronic supplementary information (ESI) available. See DOI:
10.1039/c2ob25490e
‡These authors made an equal contribution to this work.
4954 | Org. Biomol. Chem., 2012, 10, 4954–4960
This journal is © The Royal Society of Chemistry 2012

View Online
‘click’ chemistry is used to attach either cationic or PEG groups
to form the polar secondary side. This strategy offers many
possibilities for chemical functionalisation, along with relative
ease of synthesis and high yield.
In this study we synthesised two CDs, an amphiphilic cationic
CD and an amphiphilic PEGylated CD and blended these
together to yield co-formulations with various degrees of PEGy-
lation. Properties of CD : siRNA complexes formed by the cat-
ionic amphiphilic cyclodextrin alone and by the co-formulated
CDs were investigated, including size, charge, aggregation,
transfection efficiency in Caco2 cells and effects on cell
viability.
cyclodextrin 5 from 3 started with the ‘click’ reaction between 3
and tert-butyl 3-azidopropylcarbamate, carried out in water with
a cuprous catalyst to afford 4. Thioalkylation of this with
dodecane thiol and deprotection of the Boc-protected amine
groups gave 5. The PEGylated cyclodextrin 7 was also formed
from compound 3, starting with its cuprous-catalysed reaction
with ω-O-methyl-poly(ethylene glycol) azide (Scheme 2). This
reactive PEG was synthesised in two steps: monomethyl-PEG
500 was activated using mesyl chloride and the mesyl group sub-
stituted using sodium azide. The 1,3-dipolar cycloaddition to 3
was carried out in water-DMF to yield compound 6, which was
again thioalkyated using dodecane thiol, to obtain 7.
The choice of SC12 as the lipid chain at the 6-positions was
decided by previous work where we observed that self-assembly
into bilayers required chains preferably greater than C10, whereas
above C12 solubility became a problem during the synthesis.24 In
deciding on the extended PEG chain, the potential for immuno-
genicity of chains longer than PEG 500 was kept in mind.25
The use of polyethylene glycol as a surface coating for nano-
particles delivering siRNA is well established with small cationic
lipid and polymer vectors.16 PEG chains have been attached by
various means, such as chemical reaction with the particle
surface14 or post-formulation insertion into the outer layers of
nanoparticles when the PEG is part of an amphiphile.26 PEGyla-
tion of CD-based gene delivery vectors relied upon inclusion of
adamantane-PEG into the hydrophobic cavities of surface CD
groups,27–29 rather than directly attaching the PEG to the CD
itself. The rationale is that the PEG will provide a charge-neutral
and non-aggregating surface which will not interact with proteins
Results and discussion
Compound 3 was synthesised from β-cyclodextrin (Scheme 1)
and used as a common intermediate for synthesis of the cationic
and PEGylated cyclodextrins, 5 and 7. Starting from the natural
β-cyclodextrin, tert-butyldimethylsilyl chloride selectively intro-
duced the TBDMS group at the 6-positions to give 1. Reaction
with propargyl bromide yielded compound 2, which was reacted
with a triphenylphosphine bromine complex to afford the bifunc-
tional intermediate 3 with an overall yield of 55%.
CD 3 possesses two reactive chemical functions: bromo
groups for displacement by dodecyl thiolate to form the lipidic
groups of the final amphiphilic cyclodextrins and alkyne func-
tionality for “click” chemical introduction of the polar groups or
PEG groups on the 2-positions. Synthesis of the cationic
Scheme 1 Synthesis of the cationic cyclodextrin 5 (i) TBDMSCl, pyridine, 88% (ii) propargyl bromide, NaH, DMF (iii) PPh3, Br2, CH2Cl2, (iv)
tert-butyl 3-azidopropylcarbamate, CuSO4·5H2O, sodium ascorbate, DMF–H2O, 80 °C (v) (a) dodecanethiol, NaH, 80 °C, (b) TFA, CH2Cl2.
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 4954–4960 | 4955

View Online
Scheme 2 Synthesis of the pegylated cyclodextrin 7 (i) 1-azido-1-deoxy-ω-methoxy-PEG500, CuSO4, 5H2O, sodium ascorbate, DMF–H2O, 80 °C
(ii) dodecanethiol, NaH, 80 °C.
and which has scope for receptor–ligand attachment. Here we
have taken another approach, namely the co-formulation of two
structurally similar vectors, where one, CD 5, is the cationic con-
densing agent for the siRNA, and the other, CD 7, bears the
PEG chains. Their structural similarity made it possible to apply
them as a dual formulation. This strategy enables numerous pos-
sibilities for varying the formulation by, for example, attaching
different length PEG chains to the PEGylated CD, attaching cell-
specific targeting ligands via the ‘click’ linker or by adjusting
the proportions of the two CDs incorporated into the CD co-for-
mulation. A similar approach was reported for a polyamido-
amine-based polymer which was mixed with its PEGylated
counterpart, before siRNA complexation to yield polyplexes
with increasing extent of PEGylation.30
In this paper, firstly the cationic CD 5 was assessed, after com-
plexation with siRNA at increasing mass ratios (MR, μg CD
5 : μg siRNA). Then CD 5 was co-formulated with CD 7, at
molar ratio CD 5 : CD 7 1 : 0 to 1 : 1, yielding a series of CD for-
mulations with varying degrees of PEGylation, which were sub-
sequently complexed with siRNA. The properties and
transfection efficiency of all formulations were assessed.
The siRNA binding efficiency of the cationic CD 5 was
measured using agarose gel electrophoresis. Uncomplexed
siRNA, used as a control, migrated freely through the gel. At a
MR of 2, almost no migration was observed and at MR 5 and
above there was complete binding (results not shown). Mixing
the PEGylated CD 7 with the cationic CD up to equimolar ratio,
while maintaining the same cationic CD : siRNA mass ratio, did
not affect complexation. This observation is in contrast to a pre-
vious study which showed lack of pDNA condensation when a
CD-based polymer was PEGylated before complexation29 and
another whereby a co-formulation consisting of cationic and
neutral amphiphilic CDs exhibited poor protection of pDNA.36
CD : siRNA complex nanoparticles were characterised in
terms of size and charge, both of which can influence transfec-
tion efficiency.31,32 Upon complexation with siRNA at MR 10
(equivalent to N : P ratio 5.9), the cationic CD 5 (1 : 0) gave
nanoparticles of ∼240 nm (Fig. 1). Smaller particles (<150 nm)
were obtained at MR 20 and MR 30 (data not shown). A
reduction in particle size was observed upon addition of the
PEGylated CD to the formulation, in particular at molar ratio
1 : 1 (Fig. 1). However, the polydispersity index at this molar
ratio was 0.6, indicating considerable variation in particle size.
Fig. 1 Effects of molar ratio of cationic CD 5 to pegylated CD 7 on
the size (Z-Ave) and charge (zeta potential) of CD : siRNA complexes
(mass ratio cationic CD 5 : siRNA = 10) in water. Results are expressed
as Mean S.D. (n = 5).
The complexes formed between cationic CD 5 and siRNA
had a positive charge (+42 mV) at MR 10 (Fig. 1). A positive
surface charge contributes to interaction with cellular membrane
proteoglycans, enabling uptake.9 The co-formulations produced
by mixing the PEGylated CD with the cationic CD had reduced
surface charges (+19 mV at molar ratio 1 : 1), an expected shield-
ing effect by the PEG groups at the particle surface. A similar
modest effect on charge has previously been reported after
inclusion of PEG groups into polyethylenimine and cyclodex-
trin-based polymers.15 Masking the surface charge of cationic
nanoparticles has been reported to reduce toxicity13,30 and mini-
mise non-specific interactions with plasma components.30,33
As well as reducing surface charge of the nanoparticles, incor-
poration of the PEGylated CD had a significant effect on aggre-
gation in OptiMEM transfection medium (Fig. 2). Complexes
formed with the cationic CD 5 and siRNA (MR10) aggregated
in the high salt environment of the medium, whereas aggregation
was completely prevented in the formulations containing the
highest proportions of the PEGylated CD 7 (molar ratio 10 : 1
and 1 : 1). Therefore stability of the formulation was improved
by incorporation of the PEGylated CD 7.
Gene silencing by cationic CD 5 in Caco2 cells was assessed
by ability of the CD : siRNA complexes to silence an exogen-
ously transfected luciferase reporter plasmid. As shown in
Fig. 3, the optimal mass ratio for gene silencing was considered
to be MR 10, at which a significant reduction in luciferase gene
4956 | Org. Biomol. Chem., 2012, 10, 4954–4960
This journal is © The Royal Society of Chemistry 2012

View Online
Fig. 2 Effect of molar ratio of cationic CD 5 to pegylated CD 7 on
size (Z-Ave) of CD : siRNA complexes in deionised water and
OptiMEM (mass ratio cationic CD 5 : siRNA = 10). Results are
expressed as Mean S.D. (n = 5).
Fig. 4 Effects of molar ratio of cationic CD 5 to pegylated CD 7 on
silencing of luciferase reporter gene expression in Caco2 cells at mass
ratio cationic CD 5 : siRNA = 10. Results are expressed as Mean S.E.
M. (n = 3). *p < 0.05 relative to untreated cells.
Fig. 3 Silencing of luciferase reporter gene expression in Caco2 cells
with cationic CD 5 at increasing mass ratios (MR, cationic CD
5 : siRNA). Lipofectamine 2000 (Lf 2000) was included as a positive
control. Results are expressed as Mean S.E.M. (n = 3). *p < 0.05 rela-
tive to untreated cells.
Fig. 5 Effect of cationic CD 5 (solid line) and pegylated CD 7 (dotted
line) on viability of Caco2 cells over concentration range 0 to 60 μM.
Cells were treated for 4 h and cell viability was assessed by an MTT
assay. Results are expressed as Mean S.D. (n = 3).
expression (76
7%, *p < 0.05 compared to controls) was
observed. No further improvement in knockdown was observed
at MR 20 (70 8%). A similar level of knockdown was achieved
with the cationic lipid vector LipofectamineTM 2000. Neither
siRNA alone, nor a non-silencing control siRNA complexed to
CD 5 (MR 10 : non-silencing siRNA (MR10/ns)), mediated gene
silencing (Fig. 3). Due to its high transfection efficiency, MR 10
was chosen for further experiments using the mixed CD
formulation.
On addition of the PEGylated CD, no significant knockdown
was observed even with low proportions incorporated into the
formulation (Fig. 4). This was despite the fact that the PEGylated
CD 7 had no effect on siRNA binding and did not make the co-
formulated complexes totally charge-neutral. It is possible that
steric effects of the PEG groups blocked charge-interaction with
the cell membranes, or that the PEG groups interfered with intra-
cellular trafficking. PEGylation of particles formed with DNA
has previously been shown to reduce the transfection efficiency
of polyethylenimine (PEI) and a linear cyclodextrin-containing
polymer,15 an effect attributed to either reduced cellular uptake
or impaired intracellular processing. Similarly, a PEGylated lipid
vector was unable to deliver pDNA or siRNA in vitro as
compared to other polymeric or lipid vectors.34 However, in con-
trast to these results, another study showed improved gene silen-
cing in vitro with a PEG-modified PEI vector,14 due to various
factors including improved siRNA stability in serum and effec-
tive endosomal release. When PEG was conjugated to poly-L-
lysine by a pH-sensitive linker, it was considered to be cleaved
in the acidic environment of the endosome so that it did not
interfere with subsequent siRNA-mediated gene silencing.13
Therefore, the effects of PEGylation on transfection appear to
depend on linker and environment. Also, reduction of transfec-
tion in vitro may not be seen in vivo, as was reported by Hata-
keyama et al. who observed greater activity with a PEGylated
lipid vector on systemic administration at a tumour site.35
Finally, the cytotoxicity of each of the CD vectors in Caco2
cell cultures was assessed by means of an MTT assay. Cells were
incubated with increasing concentrations of either cationic CD 5
or 7 for four hours. For the cationic CD, at concentrations of up
to 30 μM, at least 85% of cells remained viable compared to
untreated controls (Fig. 5). The concentration of cationic CD
used for transfection experiments at CD : siRNA MR10 was
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 4954–4960 | 4957

View Online
approximately 1.7 μM – little or no cytotoxicity was evident at
this concentration. The PEGylated CD did not affect viability at
any of these concentrations, as expected in view of its neutral
charge.
1H NMR (500 MHz, CDCl3) δ 4.94 (d, J = 3.6 Hz, 1H), 4.66
(s, 1H), 4.50 (t, J = 2.4 Hz, 2H), 3.98–3.87 (m, 2H), 3.63 (d, J =
11.2 Hz, 1H), 3.52 (dd, J = 9.0, 4.3 Hz, 3H), 2.42 (t, J = 2.3 Hz,
1H), 0.84 (s, 9H), 0.00 (d, J = 2.7 Hz, 6H). 13C NMR
(126 MHz, CDCl3) δ 101.32, 82.13, 79.69, 79.12, 75.20, 73.62,
71.92, 61.92, 59.56, 27.13, 26.13, 18.49, −4.93. MALDI-MS:
2224.48 [M + Na]+; 2240.58 [M + K]+.
Conclusions
Copper-catalysed ‘click’ chemistry can be used to modify the
2′-hydroxyl position of β-cyclodextrin in the synthesis of amphi-
philic CDs having either cationic primary amine groups or
lengthy PEG chains on this secondary side. The two types of
amphiphile are compatible for co-formulation into nanoparticles
with suitable properties for siRNA delivery.
High levels of gene silencing were achieved in vitro with a
cationic CD alone. Nanoparticles formed with siRNA using dual
formulation of the cationic and a PEGylated CD displayed
improved physicochemical properties, namely reduced surface
charge and lowered aggregation. As expected, the modified par-
ticle surface also led to reduced transfection efficiency compared
with the unPEGylated particles, probably due to steric masking
of the positive charge by the PEG groups. However, the pro-
vision of receptor cell-targeting ligands at this surface is
expected to restore transfection efficiency, but with the added
advantages of specificity and lowered toxicity. Meanwhile this
approach to formulating PEGylated vector-siRNA complexes is
a further demonstration of the advantages of cyclodextrins as
core molecules for siRNA delivery.
Heptakis(6-bromo-6-deoxy-2-O-propargyl)-β-cyclodextrin (3).
Triphenylphosphine (1.22 g, 4.0 mmol, 22 eq.) in 50 mL of
anhydrous dichloromethane was stirred at 0 °C under nitrogen
for 30 min before addition of bromine (180 μL, 3.63 mmol, 20
eq.) dropwise. The solution was stirred for 30 min at 0 °C and
for 2 h at r.t. until a white precipitate appeared. The solution was
then cooled to 0 °C and 400 mg (0.18 mmol, 1 eq.) of heptakis
(2-O-propargyl-6-O-tert-butyldimethylsilyl)-β-cyclodextrin
2
was added. The solution was stirred at r.t. for 12 h, then concen-
trated under reduced pressure. The residue was taken into ethyl
acetate (50 mL) and washed with brine (100 mL). The residue
after re-evaporation was purified by column chromatography
over SiO2 (cyclohexane–ethyl acetate, 4 : 1 to 1 : 1) to give 3 as
a colourless solid (296 mg, 88%).
1H NMR (500 MHz, CDCl3) δ 5.07 (d, J = 3.8 Hz, 1H), 4.76
(s, 1H), 4.55 (d, J = 2.3 Hz, 2H), 3.97 (t, J = 9.2 Hz, 1H),
3.88–3.82 (m, 2H), 3.75–3.67 (m, 2H), 3.41 (t, J = 9.2 Hz, 1H),
2.53 (t, J = 2.3 Hz, 1H). 13C NMR (126 MHz, CDCl3) δ 101.74,
85.93, 79.21, 78.40, 75.93, 73.11, 70.84, 59.85, 33.12.
MALDI-MS: 1863.63 [M + Na]+; 1879.84 [M + K]+.
Heptakis[6-bromo-2-O-(N-(N-t-butoxycarbonyl-3′′-aminopro-
pyl)-1′H-triazole-4′-yl-methyl)]-β-cyclodextrin (4). Compound 3
(420 mg, 0.23 mmol, 1 eq.) and tert-butyl 3-azidopropylcarba-
mate (410 mg, 2.05 mmol, 9 eq.) were dissolved in 20 mL of
DMF and 20 mL of water. Copper(II) sulfate pentahydrate
(45.54 mg, 0.18 mmol, 0.8 eq.) was added, then 112 mg
(0.57 mmol, 2.5 eq.) of sodium ascorbate was added and the sol-
ution was stirred at 80 °C for 12 h. The solution was concen-
trated under reduced pressure and purified by column
chromatography over SiO2 (dichloromethane–methanol 9 : 1) to
give product (715 mg, 96%).
Experimental details
Materials
Cyclodextrin (Aldrich) was dried for 12 h at 100 °C under
vacuum. tert-Butyl 3-azidopropylcarbamate was from Lumino-
Chem, and for safety this low-MW azide was stored as a refriger-
ated solution in dimethylformamide. Monomethoxy PEG500
was from Aldrich. Anti-luciferase siRNA (sense strand CUU
ACG AGUA CU UCG AdT dT, antisense strand UCG AAG
UAC UCA GCG UAA GdT dT), and negative siRNA (sense
strand UUC UCC GAA CGU GUC ACG UdT dt) were from
Qiagen.
1H NMR (400 MHz, CDCl3) δ 7.82 (s, 1H), 5.34 (s, 1H),
5.04–4.81 (m, 3H), 4.43 (t, J = 6.6 Hz, 2H), 3.87 (t, J = 9.0 Hz,
1H), 3.74 (t, J = 8.6 Hz, 2H), 3.64–3.47 (m, 2H), 3.24 (t, J = 8.9
Hz, 1H), 3.09 (dd, J = 12.0, 5.9 Hz, 2H), 2.12–2.00 (m, 2H),
1.37 (s, 9H). 13C NMR (101 MHz, CDCl3) δ 156.23, 143.83,
123.91, 101.17, 85.50, 79.27, 79.02, 72.71, 70.63, 65.35, 47.76,
37.33, 32.92, 30.73, 28.37.
Synthesis of modified cyclodextrins
Heptakis(2-O-propargyl-6-O-tert-butyldimethylsilyl)-β-cyclo-
dextrin (2). Heptakis(6-O-tert-butyldimethylsilyl)-β-cyclodex-
trin 1 (1 g, 0.52 mmol, 1 eq.) in 100 mL of anhydrous DMF was
stirred at 0 °C under nitrogen for 30 min. Sodium hydride
(125 mg, 5.17 mmol, 10 eq.) was added slowly, and the solution
was stirred at 0 °C for 3 h and at r.t. for 12 h. The solution was
then cooled to 0 °C and 550 mg (4.65 mmol, 9 eq.) of propargyl
bromide (80% in toluene) was added. The resulting solution was
stirred for 2 h at 0 °C and 12 h at r.t. under N2. The solution was
concentrated under reduced pressure, taken into ethyl acetate
(50 mL) and washed with brine (100 mL). The residue after
evaporation was purified by column chromatography over SiO2
(cyclohexane–ethyl acetate, 3 : 1) to give 2 as a colourless solid
(765 mg, 68%).
Heptakis[2-O-(N-(N-t-butoxycarbonyl-3′′-aminopropyl)-1′H-
triazole-4′-yl-methyl)-6-dodecylthio]-β-cyclodextrin. Dodecanethiol
(978.7 mg, 4.84 mmol, 22 eq.) and 116.3 mg (4.84 mmol, 22
eq.) of sodium hydride were solubilised in 50 mL of anhydrous
DMF. The solution was stirred at 80 °C under N2 for 2 h. The
solution was then cooled to room temperature and 715.1 mg
(0.22 mmol, 1 eq.) of 4 in 5 mL of anhydrous DMF was added.
The solution was stirred overnight at 80 °C under N2. The sol-
ution was concentrated under reduced pressure and purified by
column chromatography over SiO2 (dichloromethane–methanol
9 : 1) to give product (649 mg, 72%).
4958 | Org. Biomol. Chem., 2012, 10, 4954–4960
This journal is © The Royal Society of Chemistry 2012

View Online
1H NMR (400 MHz, CDCl3) δ 7.85 (s, 1H), 5.34 (s, 1H),
5.05 (d, J = 12.1 Hz, 1H), 4.90 (d, J = 11.8 Hz, 2H), 4.48 (t, J =
6.6 Hz, 2H), 3.89 (t, J = 9.1 Hz, 1H), 3.85–3.75 (m, 1H), 3.51
(d, J = 8.0 Hz, 1H), 3.43 (t, J = 9.0 Hz, 1H), 3.16 (dd, J = 12.1,
6.0 Hz, 2H), 3.00–2.82 (m, 2H), 2.56 (t, J = 7.4 Hz, 2H),
2.21–2.06 (m, 2H), 1.61–1.17 (m, 29H), 0.88 (t, J = 6.8 Hz,
3H). 13C NMR (101 MHz, CDCl3) δ 156.24, 144.20, 123.77,
101.09, 85.50, 79.28, 73.02, 71.05, 65.59, 47.86, 37.45, 33.74,
31.96, 30.81, 30.57–29.39 (m, 9C), 29.13, 28.45, 22.71, 14.12.
Heptakis[6-bromo-2-O-(N-(1′′-deoxy-ω-methoxy-PEG500-1′′-yl)-
1′H-triazole-4′-yl-methyl)]-β-cyclodextrin (6). Compound
3
(100 mg, 0.054 mmol, 1 eq.) and 1-azido-1-deoxy-ω-methoxy-
PEG500 (317 mg, 0.54 mmol, 10 eq.) were dissolved in 6 mL of
DMF and 6 mL of water. 13.5 mg (0.054 mmol, 1 eq.) of copper
(II) sulfate pentahydrate was added. 32 mg (0.16 mmol, 3 eq.) of
sodium ascorbate was added and the solution was stirred at
80 °C for 12 h. The solution was concentrated under reduced
pressure and purified by size exclusion column chromatography
over SEPHADEX LH20 (methanol 100%) to give product
(282 mg, 88%).
Heptakis[2-O-(N-(3′′-aminopropyl)-1′H-triazole-4′-yl-methyl)-
6-dodecylthio]-β-cyclodextrin trifluoroacetate (5). Heptakis[2-O-
(N-(N-t-butoxycarbonyl-3′′-aminopropyl)-1′H-triazole-4′-yl-
methyl)-6-dodecylthio]-β-cyclodextrin (649 mg, 0.16 mmol)
was dissolved in 20 mL of DCM and 20 mL of TFA. The sol-
ution was stirred at r.t. under N2 for 6 h. The solution was then
concentrated under reduced pressure and dried under high
vacuum to give product (628 mg, 95%).
1H NMR (400 MHz, CDCl3) δ 7.90 (s, 1H), 5.07–4.89 (m,
3H), 4.84 (s, 1H), 4.69–4.47 (m, 1H), 4.01–3.23 (m, 52H). 13C
NMR (101 MHz, CDCl3) δ 143.62, 124.66, 101.35, 85.54,
78.41, 73.15–69.03 (m, 23), 64.90, 58.97, 50.25, 33.01.
Heptakis[2-O-(N-(1′′-deoxy-ω-methoxy-PEG500-1′′-yl)-1′H-
triazole-4′-yl-methyl)-6-dodecylthio]-β-cyclodextrin (7). Dode-
canethiol (62.24 mg, 0.308 mmol, 15 eq.) and sodium hydride
(8 g, 0.35 mmol, 17 eq.) were dissolved in 5 mL of anhydrous
DMF. The solution was stirred at 80 °C under N2 for 2 h. The
solution was then cooled to room temperature and 114 mg
(0.021 mmol, 1 eq.) of 6 in 5 mL of anhydrous DMF was added.
The solution was stirred overnight at 80 °C under N2. The sol-
ution was concentrated under reduced pressure and purified by
size exclusion column chromatography over SEPHADEX LH20
(methanol 100%) to give product (130 mg, quant.).
1H NMR (400 MHz, CD3OD–CDCl3) δ 7.98 (s, 1H), 5.05 (s,
1H), 4.92 (dd, J = 54.1, 11.7 Hz, 2H), 4.56 (t, J = 6.6 Hz, 2H),
3.83 (t, J = 8.8 Hz, 2H), 3.59–3.45 (m, 2H), 3.08–2.86 (m, 4H),
2.61 (t, J = 7.0 Hz, 2H), 2.39–2.28 (m, 2H), 1.64–1.53 (m, 2H),
1.43–1.20 (m, 18H), 0.89 (t, J = 6.7 Hz, 3H). 13C NMR
(101 MHz, CD3OD–CDCl3) δ 144.47, 124.85, 101.13, 85.75,
80.63, 73.25, 71.57, 65.60, 47.62, 37.11, 34.11, 33.90, 32.31,
30.63–29.19 (m, 9C), 28.15, 23.03, 14.33. MALDI-MS:
3394.55 [M + H]+; 3416.28 [M + Na]+. λmax 220 nm. FT-IR
(cm−1): 3414, 2923, 2853, 1133, 1048.
1H NMR (300 MHz, CDCl3) δ 7.81 (s, 1H), 5.07–4.83 (m,
2H), 4.78 (s, 1H), 4.63–4.44 (m, 2H), 4.04–3.22 (m, 50H),
3.06–2.68 (m, 2H), 2.68–2.38 (m, 2H), 1.64–1.40 (m, 2H),
1.40–1.11 (m, 20H), 0.84 (t, J = 6.5 Hz, 3H). 13C NMR
(75 MHz, CDCl3) δ 143.95, 124.49, 101.20, 85.50, 79.05,
74.13–68.27 (m, 24C), 65.04, 58.98, 50.14, 33.64 (2C), 31.90,
30.89–28.25 (m, 8C), 22.65, 14.07. MALDI–MS: Mavg = 6350 g
mol−1. λmax 210 nm. FT-IR (cm−1): 3422, 2922, 2855, 1107, 1047.
1-O-Mesyl-ω-methoxy-PEG500. Monomethyl-PEG 500 (1.42 g,
2.57 mmol, 1 eq.) in 80 mL of DCM and 3.2 mL (18.02 mmol,
7 eq.) of diisopropylethylamine was cooled to 0 °C and 1 mL
(12.87 mmol, 5 eq.) of mesyl chloride was added. The solution
was stirred for 12 h at r.t. under N2. The solution was concen-
trated under reduced pressure and the residue was purified by
column chromatography over SiO2 (CH2Cl2–methanol, 95 : 5) to
give product (1.52 g, 94%).
1H NMR (500 MHz, CDCl3) δ 4.27–4.22 (m, 2H), 3.66–3.62
(m, 2H), 3.62–3.45 (m, 40H), 3.43–3.40 (m, 2H), 3.24 (s, 3H),
Preparation and characterisation of CD:siRNA complexes
Cyclodextrins, dissolved in chloroform (1 mg ml−1), were mixed
to give required molar ratios. The solvent was removed with a
gentle stream of nitrogen. Aliquots were stored at −20 °C until
required. For preparation of CD:siRNA complexes, CDs were
2.96 (s, 3H). 13C NMR (126 MHz, CDCl3)
δ 71.70,
70.53–69.85 (m), 69.19, 68.77, 58.72, 37.45. MS (ESI−/TOF)
m/z: 506.73 (16%) [M − H]−; 551.16 (57%) [M − H]−; 595.28
(73%) [M − H]−; 639.13 (81%) [M − H]−; 683.30 (100%) [M
− H]−; 727.29 (96%) [M − H]−; 771.21 (65%) [M − H]−;
859.40 (23%) [M − H]−; 903.41 (12%) [M − H]−.
rehydrated with deionised water (final concentration 1 μg μl−1
)
and sonicated for 1 h at room temperature (RT), then mixed with
siRNA in an equal volume of water and used after 20–30 min.
For experiments involving the cationic and pegylated CD mixes,
a fixed cationic CD : siRNA mass ratio of 10 was chosen.
Particle size (Z-Ave) and surface charge (ζ-potential) of the
CD:siRNA complexes were measured by dynamic light scatter-
ing, using the Malvern Zetasizer Nano. Samples were measured
in deionised water and OptiMEM (Invitrogen) (final volume
1 ml containing 3 μg siRNA), in two independent experiments,
each with five measurements.
For measurement of cyclodextrin–siRNA binding by gel elec-
trophoresis, the cyclodextrin and siRNA were mixed with
loading buffer and deionised water to a final volume of 20 μl
(containing 0.3 μg siRNA). Samples were added to wells in a
1% agarose gel containing ethidium bromide (0.5 μg ml−1).
Electrophoresis was carried out at 90 V for 20 min, with a Tris-
1-Azido-1-deoxy-ω-methoxy-PEG500. 1-O-Mesyl-ω-methoxy-
PEG500 (1.52 g, 2.38 mmol, 1 eq.) was dissolved in 30 mL of
DMF. 620 mg (9.55 mmol, 4 eq.) of sodium azide was added.
The solution was heated at 80 °C overnight. The solution was
concentrated under reduced pressure and the residue was purified
by column chromatography over SiO2 (CH2Cl2–methanol,
95 : 5) to give product (1.21 g, 87%).
1H NMR (400 MHz, CDCl3) δ 3.70–3.50 (m, 48H),
3.50–3.44 (m, 2H), 3.34–3.28 (m, 5H). 13C NMR (101 MHz,
CDCl3) δ 71.85, 70.91–70.18 (m), 69.95, 58.93, 50.60. MS
(ESI+/TOF) m/z: 476.15 (15%) [M + Na]+; 520.14 (51%)
[M + Na]+; 564.14 (100%) [M + Na]+; 608.19 (93%) [M + Na]+;
652.18 (62%) [M + Na]+; 696.24 (26%) [M + Na]+; 740.29
(15%) [M + Na]+.
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 4954–4960 | 4959

View Online
borate-EDTA buffer. Bands corresponding to the DNA ladder
(100 b.p.) and unbound siRNA were visualised by UV, using the
DNR Bioimaging Systems MiniBis Pro and Gel Capture US B2
software.
References
1 S. M. Elbashir, J. Harborth, W. Lendeckel, A. Yalcin, K. Weber and
T. Tuschl, Nature, 2001, 411, 494–498.
2 D. Bumcrot, M. Manoharan, V. Koteliansky and D. W. Y. Sah, Nat.
Chem. Biol., 2006, 2, 711–719.
3 S. A. Cryan, R. Donohue, B. J. Ravo, R. Darcy and C. M. O’Driscoll, J.
Drug Delivery Sci. Technol., 2004, 14, 57–62.
Cell culture and transfection
4 S. A. Cryan, A. Holohan, R. Donohue, R. Darcy and C. M. O’Driscoll,
Eur. J. Pharm. Sci., 2004, 21, 625–633.
5 C. O. Mellet, J. M. G. Fernandez and J. M. Benito, Chem. Soc. Rev.,
2011, 40, 1586–1608.
6 C. Byrne, F. Sallas, D. K. Rai, J. Ogier and R. Darcy, Org. Biomol.
Chem., 2009, 7, 3763–3771.
7 A. Diaz-Moscoso, D. Vercauteren, J. Rejman, J. M. Benito, C. O. Mellet,
S. C. De Smedt and J. M. G. Fernandez, J. Controlled Release, 2010,
143, 318–325.
8 A. McMahon, E. Gomez, R. Donohue, D. Forde, R. Darcy and C.
M. O’Driscoll, J. Drug Delivery Sci. Technol., 2008, 18, 303–307.
9 M. J. O’Neill, J. Guo, C. Byrne, R. Darcy and C. M. O’Driscoll,
Int. J. Pharm., 2011, 413, 174–183.
10 A. Méndez-Ardoy, K. Urbiola, C. Aranda, C. Ortiz-Mellet, J. M. García-
Fernández and C. Tros de Ilarduya, Nanomedicine, 2011, 6, 1697–1707.
11 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.
12 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.
13 J. Guo, W. P. Cheng, J. Gu, C. Ding, X. Qu, Z. Yang and C. O’Driscoll,
Eur. J. Pharm. Sci., 2012, 45, 521–532.
Caco2 (human colon carcinoma) cells were cultured in Dulbec-
co’s modified Eagle’s medium (DMEM) containing 10% fetal
bovine serum (FBS) and L-glutamine (all from Sigma-Aldrich),
in a 37 °C, 5% CO2 environment. For transfection, cells were
seeded at 1 × 105 cells per well in 24-well plates, or 3.5 × 104
cells in 96-well plates. Transfection was carried out when cul-
tures were ∼70% confluent.
For gene silencing, Caco2 cell cultures were transfected with
luciferase reporter plasmid pGL3 luciferase (1 μg per well) com-
plexed with LipofectamineTM 2000 (Invitrogen) (2.5 μl μg−1
pDNA), for 2 h. Following this, cultures were washed twice with
phosphate-buffered saline (PBS, Sigma-Aldrich). siRNA (50 nM)
alone, or complexed with LipofectamineTM 2000 or CD(s), in
OptiMEM (Invitrogen) was added to the cultures in serum-con-
taining medium. At 4 h media were replaced, and a further 20 h
later cells were washed with PBS, then lysed with 400 μl of 1×
Reporter Lysis Buffer (Promega Corp.) and frozen at −80 °C.
Lysate was collected and centrifuged for 5 min at 13 000 rpm. A
sample of the supernatant (20 μl) was assayed for expression of
luciferase by adding to 100 μl of luciferin (Promega), and
measuring immediately in a luminometer. Luciferase expression
is reported in relative luminescence units (RLU) per mg protein,
as determined by BCA Protein Assay (Pierce). Results are
expressed as Mean S.E.M. of triplicate values.
14 S. Mao, M. Neu, O. Germershaus, O. Merkel, J. Sitterberg, U. Bakowsky
and T. Kissel, Bioconjugate Chem., 2006, 17, 1209–1218.
15 S. Mishra, P. Webster and M. E. Davis, Eur. J. Cell Biol., 2004, 83, 97.
16 K. Kostarelos and A. D. Miller, Chem. Soc. Rev., 2005, 34, 970–994.
17 A. Méndez-Ardoy, N. Guilloteau, C. Di Giorgio, P. Vierling, F. Santoyo-
González, C. Ortiz Mellet and J. M. García Fernández, J. Org. Chem.,
2011, 76, 5882–5894.
18 S. Srinivasachari, K. M. Fichter and T. M. Reineke, J. Am. Chem. Soc.,
2008, 130, 4618–4627.
19 S. Srinivasachari and T. M. Reineke, Biomaterials, 2009, 30, 928–938.
20 A. Mendez-Ardoy, M. Gomez-Garcia, C. O. Mellet, N. Sevillano, M.
D. Giron, R. Salto, F. Santoyo-Gonzalez and J. M. G. Fernandez, Org.
Biomol. Chem., 2009, 7, 2681–2684.
21 S. Ward and C. C. Ling, Eur. J. Org. Chem., 2011, 4853–4861.
22 F. Sallas and R. Darcy, Eur. J. Org. Chem., 2008, 957–969.
23 R. Donohue, A. Mazzaglia, B. J. Ravoo and R. Darcy, Chem. Commun.,
2002, 2864–2865.
24 A. Mazzaglia, R. Donohue, B. J. Ravoo and R. Darcy, Eur. J. Org.
Chem., 2001, 2001, 1715–1721.
Cell viability in presence of cyclodextrins was determined by
the MTT assay, which measures the reduction of MTT (3-(4,5-
dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) by mito-
chondrial dehydrogenase in viable cells. Cell cultures were
treated with CD 5 or 7 for 4 h. Media was replaced with 100 μl
fresh media and MTT (20 μl of a 5 mg ml−1 solution) for four
hours, after which the formazon crystals produced were dis-
solved in 100 μl DMSO. Absorbance was measured at 590 nm
using a UV plate reader. Cell viability was calculated as a per-
centage of untreated control cells. Results are expressed as Mean
S.D. of triplicate values.
25 T. L. Cheng, P. Y. Wu, M. F. Wu, J. W. Chern and S. R. Roffler, Bioconju-
gate Chem., 1999, 10, 520–528.
26 L. S. Mendonca, F. Firmino, J. N. Moreira, M. C. P. de Lima and
S. Simoes, Bioconjugate Chem., 2010, 21, 157–168.
27 N. C. Bellocq, S. H. Pun, G. S. Jensen and M. E. Davis, Bioconjugate
Chem., 2003, 14, 1122–1132.
Statistics
28 Y. Ping, C. Liu, Z. Zhang, K. L. Liu, J. Chen and J. Li, Biomaterials,
2011, 32, 8328–8341.
29 S. H. Pun and M. E. Davis, Bioconjugate Chem., 2002, 13, 630–639.
30 P. Vader, L. J. van der Aa, J. F. J. Engbersen, G. Storm and
R. M. Schiffelers, Pharm. Res., 2012, 29, 352–361.
GraphPad Prism (San Diego, CA, USA, version 5.0) was used
for statistical calculations. One-way analysis of variance
(ANOVA) was used to compare multiple groups followed by
Bonferroni’s post hoc test. Statistical significance was set at
*p < 0.05.
31 J. Rejman, V. Oberle, I. S. Zuhorn and D. Hoekstra, Biochem. J., 2004,
377, 159–169.
32 S. Resina, P. Prevot and A. R. Thierry, PLoS One, 2009, 4, 11.
33 D. W. Bartlett and M. E. Davis, Bioconjugate Chem., 2007, 18, 456–468.
34 J. Schäfer, S. Höbel, U. Bakowsky and A. Aigner, Biomaterials, 2010,
31, 6892–6900.
35 H. Hatakeyama, H. Akita, K. Kogure, M. Oishi, Y. Nagasaki, Y. Kihira,
M. Ueno, H. Kobayashi, H. Kikuchi and H. Harashima, Gene Ther.,
2007, 14, 68–77.
36 A. Díaz-Moscoso, N. Guilloteau, C. Bienvenu, A. Méndez-Ardoy,
J. L. Jiménez Blanco, J. M. Benito, L. Le Gourriérec, C. Di Giorgio,
P. Vierling, J. Defaye, C. Ortiz Mellet and J. M. García Fernández, Bio-
materials, 2011, 32, 7263–7273.
Acknowledgements
The authors wish to acknowledge Science Foundation Ireland
(Strategic Research Cluster grant no. 07/SRC/B1154), the Irish
Drug Delivery Network and the Irish Research Council for
Science, Engineering and Technology (scholarship to
A. O’Mahony) for research funding. We also wish to thank
Dr Michael Cronin, UCC, Cork for technical assistance.
4960 | Org. Biomol. Chem., 2012, 10, 4954–4960
This journal is © The Royal Society of Chemistry 2012