Polycationic β-cyclodextrin "click clusters": Monodisperse and versatile scaffolds for nucleic acid delivery

Article abstract of DOI:10.1021/ja074597v

Herein, a novel series of multivalent polycationic β-cyclodextrin "click clusters" with discrete molecular weight have been synthesized, characterized, and examined as therapeutic pDNA carriers. The materials were creatively designed based on a β-cyclodextrin core to impart a biocompatible multivalent architecture and oligoethyleneamine arms to facilitate pDNA binding, encapsulation, and cellular uptake. An acetylated-per-azido- β-cyclodextrin (4) was reacted with series of alkyne dendrons (7a-e) (containing one to five ethyleneamine units) using copper-catalyzed 1,3-dipolar cycloaddition, to form a series of click clusters (9a-e) bearing 1,2,3-triazole linkers. Gel electrophoresis experiments, dynamic light scattering, and transmission electron microscopy revealed that the macromolecules bind and compact pDNA into spherical nanoparticles in the size range of 80-130 nm. The polycations protect pDNA against nuclease degradation, where structures 9c, 9d, and 9e did not allow pDNA degradation in the presence of serum for up to 48 h. The cellular uptake profiles were evaluated in Opti-MEM and demonstrate that all the click clusters efficiently deliver Cy5-labeled pDNA into HeLa and H9c2 (2-1) cells, and compounds 9d and 9e yielded efficacy similar to that of the positive controls, Jet-PEI and Superfect. Furthermore, the luciferase gene delivery experiments revealed that the level of reporter gene expression increased with an increase in oligoethyleneamine number within the cluster arms. The cytotoxicity profiles of these materials were evaluated by protein, MTT, and LDH assays, which demonstrate that all the click clusters remain nontoxic within the expected dosage range while the positive controls, Jet PEI and Superfect, were highly cytotoxic. In particular, 9d and 9e were the most effective and promising polycationic vehicles to be further optimized for future systemic delivery experiments.

Full text of DOI:10.1021/ja074597v

Published on Web 03/14/2008
Polycationic â-Cyclodextrin “Click Clusters”: Monodisperse
and Versatile Scaffolds for Nucleic Acid Delivery
Sathya Srinivasachari, Katye M. Fichter, and Theresa M. Reineke*
Department of Chemistry, UniVersity of Cincinnati, Cincinnati, Ohio 45221-0172
Received June 22, 2007; E-mail: Theresa.Reineke@uc.edu
Abstract: Herein, a novel series of multivalent polycationic â-cyclodextrin “click clusters” with discrete
molecular weight have been synthesized, characterized, and examined as therapeutic pDNA carriers. The
materials were creatively designed based on a â-cyclodextrin core to impart a biocompatible multivalent
architecture and oligoethyleneamine arms to facilitate pDNA binding, encapsulation, and cellular uptake.
An acetylated-per-azido-â-cyclodextrin (4) was reacted with series of alkyne dendrons (7a-e) (containing
one to five ethyleneamine units) using copper-catalyzed 1,3-dipolar cycloaddition, to form a series of click
clusters (9a-e) bearing 1,2,3-triazole linkers. Gel electrophoresis experiments, dynamic light scattering,
and transmission electron microscopy revealed that the macromolecules bind and compact pDNA into
spherical nanoparticles in the size range of 80-130 nm. The polycations protect pDNA against nuclease
degradation, where structures 9c, 9d, and 9e did not allow pDNA degradation in the presence of serum for
up to 48 h. The cellular uptake profiles were evaluated in Opti-MEM and demonstrate that all the click
clusters efficiently deliver Cy5-labeled pDNA into HeLa and H9c2 (2-1) cells, and compounds 9d and 9e
yielded efficacy similar to that of the positive controls, Jet-PEI and Superfect. Furthermore, the luciferase
gene delivery experiments revealed that the level of reporter gene expression increased with an increase
in oligoethyleneamine number within the cluster arms. The cytotoxicity profiles of these materials were
evaluated by protein, MTT, and LDH assays, which demonstrate that all the click clusters remain nontoxic
within the expected dosage range while the positive controls, Jet PEI and Superfect, were highly cytotoxic.
In particular, 9d and 9e were the most effective and promising polycationic vehicles to be further optimized
for future systemic delivery experiments.
Introduction
challenge is the design and synthesis of effective, efficient, and
selective supramolecular drug delivery vehicles.4,7,8
Advancements in medical research are transforming and
inspiring the field of supramolecular chemistry.1-4 Likewise,
macromolecular drugs such as siRNA, aptamers, and several
oligonucleotide- and gene-based therapies are initiating an
exciting paradigm shift in pharmaceutical research.2-5 Develop-
ment of novel, selective, and specific nucleic acid drugs for
numerous diseases has rapidly increased, and many candidates
have high potential to be approved for medicinal use.6 To realize
the tremendous promise of these therapeutics, the central
The area of nucleic acid delivery has recently faced a rapid
upsurge of interest from researchers in a multitude of fields.
Although many creative materials reveal promising biocom-
patibility and delivery efficiency, the structures are often plagued
by being polydisperse in nature, poorly defined, and/or difficult
to characterize.9 Unfortunately, many elegant architectures will
not be clinically relevant due to the complex regulatory
requirements that must be overcome to obtain approval of new
drug candidates for human trials.3,10 As a result, the development
of discrete and well-characterized macromolecules that bind,
compact, and deliver nucleic acids in a nontoxic and effective
manner remains an immense hurdle in advancing this field
toward the clinic.4,11
(1) (a) Behr, J.-P. Acc. Chem. Res. 1993, 26, 274-278. (b) Nielsen, P. E. Mol.
Biotechnol. 2004, 26, 233-248. (c) Wagner, E. Pharm. Res. 2004, 21,
8-14. (d) Luo, D.; Saltzman, W. M. Gene Ther. 2006, 13, 585-586. (e)
Davis, M. E. Curr. Opin. Biotechnol. 2002, 13, 128-131. (f) Akhtar, S.
Gene Ther. 2006, 13, 739-740.
(2) Pan, H.; Zheng, Q.; Guo, X. J. Pept. Sci. 2007, 13, 154-163.
(3) Lee, C. C.; Mackay, J. A.; Frechet, J. M. J.; Szoka, F. C. Nat. Biotechnol.
2005, 23, 1517-1526.
(4) Boas, U.; Heegaard, P. M. H. Chem. Soc. ReV. 2004, 33, 43-63.
(5) (a) Anderson, W. F. Science 1992, 256, 808-813. (b) Katas, H.; Alpar, H.
O. J. Controlled Release 2006, 115, 216-225. (c) Que-Gewirth, N. S.;
Sullenger, B. A. Gene Ther. 2007, 14, 283-291. (d) Rossi, J. J. Gene Ther.
2006, 13, 583-584. (e) McNamara, J. O., II; Andrechek, E. R.; Wang, Y.;
Viles, K. D.; Rempel, R. E.; Gilboa, E.; Sullenger, B. A.; Giangrande, P.
H. Nat. Biotechnol. 2006, 24, 1005-1015. (f) Langer, R.; Tirrell, D. A.
Nature 2004, 428, 487-492.
(6) (a) Boussif, O.; Lezoualc’h, F.; Zanta, M. A.; Mergny, M. D.; Scherman,
D.; Demeneix, B.; Behr, J.-P. Proc. Natl. Acad. Sci. U.S.A 1995, 92, 7297-
7301. (b) Meyer, M.; Wagner, E. Hum. Gene Ther. 2006, 17, 1062-1076.
(c) Ooya, T.; Choi, H. S.; Yamashita, A.; Yui, N.; Sugaya, Y.; Kano, A.;
Maruyama, A.; Akita, H.; Ito, R.; Kogure, K.; Harashima, H. J. Am. Chem.
Soc. 2006, 128, 3852-3853.
Dendrimers and cluster compounds possess well-defined
architectures, precise molecular weights, and multivalent func-
(7) (a) Kim, T.-I.; Baek, J.-U.; Bai, C. Z.; Park, J.-S. Biomaterials 2007, 28,
2061-2067. (b) Dufe`s, C.; Uchegbu, I. F.; Scha¨etzlein, A. G. AdV. Drug
DeliVery ReV. 2005, 57, 2177-2022. (c) Svenson, S.; Tomalia, D. A. AdV.
Drug DeliVery ReV. 2005, 57, 2106-2129.
(8) Tomalia, D. A. Macromol. Symp. 1996, 101, 243-255.
(9) (a) Breunig, M.; Lungwitz, U.; Liebl, R.; Fontanari, C.; Klar, J.; Kurtz,
A.; Blunk, T.; Goepferich, A. J. Gene Med. 2005, 7, 1287-1298. (b)
Hwang, S. J.; Davis, M. E. Curr. Opin. Mol. Ther. 2001, 3, 183-191. (c)
Wightman, L.; Kircheis, R.; Ro¨ssler, V.; Carotta, S.; Ruzicka, R.; Kursa,
M.; Wagner, E. J. Gene Med. 2001, 3, 362-372.
(10) Helms, B.; Meijer, E. W. Science 2006, 313, 929-930.
(11) Duncan, R.; Izzo, L. AdV. Drug DeliVery ReV. 2005, 57, 2215-2237.
9
4618
J. AM. CHEM. SOC. 2008, 130, 4618-4627
10.1021/ja074597v CCC: $40.75 © 2008 American Chemical Society

Polycationic
â
-Cyclodextrin “Click Clusters”
A R T I C L E S
tionalization sites.12,13 These unique and useful characteristics
provide tailorable scaffolds that have motivated their develop-
ment as nucleic acid carriers. For example, polyamidoamine
(PAMAM),8,14,15 poly-L-lysine dendrimers,16 dendritic-sper-
mines,17 -polyesters,18 and -polypeptides2,19 have been widely
studied for nucleic acid delivery. Higher generations of these
structures yield effective delivery.20 However, chemically func-
tionalizing the large number of peripheral groups on these
macromolecules with ligands such as polyethylene glycol (to
decrease aggregation and nonspecific interactions), and targeting
groups in a well-defined and reproducible manner is complex
and sometimes not possible.15,21 Loss of material definition
increases polydispersity of the final delivery vehicle, generates
difficulties with structural characterization, and can cause
problems in the clinic. Toxicity of these structures has also
become a concern that has further hampered development
efforts.11,14,22
Herein, we describe our approach to achieve a family of
discrete macromolecules with versatile features to serve as drug
and nucleic acid delivery vehicles. We were inspired by the
clinical promise of dendrimers4,10 and the distinctive features
of â-cyclodextrin, which has been shown to serve as a
multivalent core in the design of glyco-clusters,23,24 -con-
jugates,25,26 -dendrimers,27 and star polymers.28 This cyclic
maltooligosaccharide is also endowed with a biocompatible
hydrophobic cup-shaped structure that can form inclusion
complexes with biologically specific guests. For example,
â-cyclodextrin is FDA approved as an adjuvant to aid solubi-
lization of hydrophobic drugs by forming inclusion complexes
with the therapeutic molecule, which increases bioavailability.29
Davis et al. and others have shown that polymers containing
â-cyclodextrin units can complex adamantane moieties func-
tionalized with polyethylene glycol (PEG) chains and cell-type
specific targeting moieties to discourage nanoparticle floccula-
tion and encourage tissue-specific nucleic acid delivery.30,31 The
present target structures were also motivated by previous work
conducted in our group.32 We have shown that polymers created
with alternating saccharide and oligoethyleneamine monomers
can effectively complex, condense, and deliver pDNA with
exceptional biocompatibility and efficacy.33,34 Unfortunately,
those polymers suffer from polydispersity. The work herein
presents the first example of extending this saccharide-oligoam-
ine design motif into a novel family of polycationic â-cyclo-
dextrin click clusters with discrete, monodisperse, and symmetric
geometries that reveal promising DNA delivery efficacy without
cytotoxicity. This design motif also reveals a prototype of readily
tailorable scaffolds with potential to be refined as clinically
viable delivery agents for a variety of applications.
In this report, the synthesis of the core moiety, acetylated-
per-azido-â-cyclodextrin (4), and a novel series of alkyne
dendrons (7a-e) containing between 0 and 4 secondary amines
are presented (Schemes 1 and 2). The target macromolecules
were assembled employing a convergent approach via the click
reaction. This allowed us to avoid subfunctionalized impurities
around the sterically hindered core that could elicit unintended
biological activity (Scheme 3).13,35,26 We discovered through
gel electrophoresis, dynamic light scattering, and transmission
electron microscopy experiments that all the cluster compounds
bind and encapsulate pDNA into nanoparticles. These nano-
particles stabilize nucleic acids and protect them from nuclease
degradation. In addition, most of the vectors promote high
cellular delivery of pDNA (similar to that of Jet-PEI and
Superfect, the positive controls) and encourage effective gene
(12) (a) Mulder, A.; Huskens, J.; Reinhoudt, D. N. Org. Biomol. Chem. 2004,
2, 3409-3424. (b) Smith, D. K. Chem. Commun. 2006, 1, 34-44. (c) Ihre,
H.; Padilla De Jesus, O. L.; Fre´chet, J. M. J. J. Am. Chem. Soc. 2001, 123,
5908-5917.
(13) Hawker, C. J.; Fre´chet, J. M. J. J. Am. Chem. Soc. 1990, 112, 7638-7647.
(14) Braun, C. S.; Vetro, J. A.; Tomalia, D. A.; Koe, G. S.; Koe, J. G.; Middaugh,
C. R. J. Pharm. Sci. 2005, 94, 423-436.
(15) Luo, D.; Haverstick, K.; Belcheva, N.; Han, E.; Saltzman, W. M.
Macromolecules 2002, 35, 3456-3462.
(16) (a) Ohsaki, M.; Okuda, T.; Wada, A.; Hirayama, T.; Niidome, T.; Aoyagi,
H. Bioconjugate Chem. 2002, 13, 510-517. (b) Yamagata, M.; Kawano,
T.; Shiba, K.; Mori, T.; Katayama, Y.; Niidome, T. Bioorg. Med. Chem.
2007, 15, 526-532. (c) Vlasov, G. P.; Korol’kov, V. I.; Pankova, G. A.;
Tarasenko, I. I.; Baranov, A. N.; Glazkov, P. B.; Kiselev, A. V.; Ostapenko,
O. V.; Lesina, E. A.; Baranov, V. S. Russ. J. Bioorg. Chem. 2004, 30,
12-20.
(17) (a) Kostiainen, M. A.; Hardy, J. G.; Smith, D. K. Angew. Chem., Int. Ed.
2005, 44, 2556-2559. (b) Hardy, J. G.; Kostiainen, M. A.; Smith, D. K.;
Gabrielson, N. P.; Pack, D. W. Bioconjugate Chem. 2006, 17, 172-178.
(18) (a) Grinstaff, M. W. Chem. Eur. J. 2002, 8, 2838-2846. (b) Gvili, K.;
Benny, O.; Danino, D.; Machluf, M. Biopolymers 2007, 85, 379-391. (c)
Wu, D.; Liu, Y.; Jiang, X.; He, C.; Goh, S. H.; Leong, K. W. Biomacro-
molecules 2006, 7, 1879-1883.
(27) (a) Ortega-Caballero, F.; Gime´nez-Mart´ınez, J. J.; Garc´ıa-Fuentes, L.; Ortiz-
Salmero´n, E.; Santoyo-Gonza´lez, F.; Vargas-Berenguel, A. J. Org. Chem.
2001, 66, 7786-7795. (b) Jayaraman, N.; Stoddart, J. F. Tetrahedron Lett.
1997, 38, 6767-6770. (c) Turnbull, W. B.; Stoddart, J. F. ReV. Mol.
Biotechnol. 2002, 90, 231-255. (d) Vargas-Berenguel, A.; Ortega-
Caballero, F.; Santoyo-Gonza´lez, F.; Garc´ıa-Lo´pez, J. J.; Gime´nez-Mart´ınez,
J. J.; Garc´ıa-Fuentes, L.; E, Ortiz.-Salmero´n, E. Chem. Eur. J. 2002, 8,
812-827.
(19) Sadler, K.; Tam, J. P. ReV. Mol. Biotechnol. 2002, 90, 195-229.
(20) (a) Gillies, E. R.; Fre´chet, J. M. J. Drug DiscoVery Today 2005, 10, 35-
43. (b) Gillies, E. R.; Fre´chet, J. M. J. J. Am. Chem. Soc. 2002, 124, 14137-
14146.
(28) (a) Hoogenboom, R.; Moore, B. C.; Schubert, U. S. Chem. Comm. 2006,
38, 4010-4012. (b) Yang, C.; Li, H.; Goh, S. H.; Li, J. Biomaterials 2007,
28, 3245-3254.
(21) (a) Petersen, H.; Fechner, P. M.; Martin, A. L.; Kunath, K.; Stolnik, S.;
Roberts, C. J.; Fischer, D.; Davies, M. C.; Kissel, T. Bioconjugate Chem.
2002, 13, 845-854. (b) Lee, J. H.; Lim, Y.-B.; Choi, J. S.; Lee, Y.; Kim,
T.-I.; Kim, H. J.; Yoon, J. K.; Kim, K.; Park, J.-S. Bioconjugate Chem.
2003, 14, 1214-1221.
(22) (a) Jevprasesphant, R.; Penny, J.; Jalal, R.; Attwood, D.; McKeown, N.
B.; D’Emanuele, A. J. Int. J. Pharm. 2003, 252, 263-266. (b) Malik, N.;
Wiwattanapatapee, R.; Klopsch, R.; Lorenz, K.; Frey, H.; Weener, J. W.;
Meijer, E. W.; Paulus, W.; Duncan, R. J. Controlled Release 2000, 65,
133-148.
(29) (a) Mocanu, G.; Vizitiu, D.; Carpov, A. J. Bioact. Compat. Pol. 2001, 16,
315-342. (b) Cryan, S.-A.; Holohan, A.; Sonohue, R.; Darcy, R.;
O’Driscoll, C. M. Eur. J. Pharm. Sci. 2004, 21, 625-633.
(30) (a) Hwang, S. J.; Bellocq, N. C.; Davis, M. E. Bioconjugate Chem. 2001,
12, 280-290. (b) Reineke, T. M.; Davis, M. E. Bioconjugate Chem. 2003,
14, 247-254.
(31) Bellocq, N. C.; Pun, S. H.; Jensen, G. S.; Davis, M. E. Bioconjugate Chem.
2003, 14, 1122-1132.
(32) (a) Liu, Y.; Reineke, T. M. J. Am. Chem. Soc. 2005, 127, 3004-3015. (b)
Liu, Y. W. L.; Lynch, M.; Reineke, T. M. In Polymeric Drug DeliVery
Volume I - Particulate Drug Carriers; Svenson, S., Ed.; American Chemical
Society: Washington, DC, 2005; Vol. 923, pp 217-227. (c) Srinivasachari,
S.; Liu, Y.; Zhang, G.; Prevette, L.; Reineke, T. M. J. Am. Chem. Soc.
2006, 128, 8176-8184. (d) Prevette, L. E.; Kodger, T. E.; Reineke, T. M.;
Lynch, M. E. Langmuir 2007, 23, 9773-9784.
(23) (a) Andre´, S.; Kaltner, H.; Furuike, T.; Nishimura, S.-I.; Gabius, H.-J.
Bioconjugate Chem. 2004, 15, 87-98. (b) Furuike, T.; Sadamoto, R.;
Niikura, K.; Monde, K.; Sakairi, N.; Nishimura, S. -I. Tetrahedron 2005,
61, 1737-1742. (c) Ortiz-Mellet, C.; Benito, J. M.; Garc´ıa Ferna´ndez, J.
M.; Law, H.; Chmurski, K.; Defaye, J.; O’Sullivan, M. L.; Caro, H. N.
Chem. Eur. J. 1998, 4, 2523-2531. (d) Fulton, D. A.; Stoddart, J. F. Org.
Lett. 2000, 2, 1113-1116.
(33) Liu, Y.; Reineke, T. M. Bioconjugate Chem. 2006, 17, 101-108.
(34) Liu, Y.; Reineke, T. M. Bioconjugate Chem. 2007, 18, 19-30.
(35) (a) Calvo-Flores, F. G.; Isac-Garc´ıa, J.; Herna´ndez-Mateo, F.; Pe´rez-
Balderas, F.; Calvo-As´ın, J. A.; Sanche´z-Vaquero, E.; Santoyo-Gonza´lez,
F. Org. Lett. 2000, 2, 2499-2502. (b) Yoon, K.; Goyal, P.; Weck, M.
Org. Lett. 2007, 9, 2051-2054. (c) Lee, J. W.; Kim, J. H.; Kim, H. J.;
Han, S. C.; Kim, J. H.; Shin, W. S.; Jin, S.-H. Bioconjugate Chem. 2007,
18, 579-584. (d) Binder, W. H.; Sachsenhofer, R. Macromol. Rapid
Commun. 2007, 28, 15-54.
(24) Boger, J.; Corcoran, R. J.; Lehn, J.-M. HelV. Chim. Acta 1978, 61, 2190-
2218.
(25) Garc´ıa-Lo´pez, J. J.; Herna´ndez-Mateo, F.; Isac-Garc´ıa, J.; Kim, J. M.; Roy,
R.; Santoyo-Gonza´lez, F.; Vargas-Berenguel, A. J. Org. Chem. 1999, 64,
522-531.
(26) Pe´rez-Balderas, F.; Ortega-Mun˜oz, M.; Morales-Sanfrutos, J.; Herna´ndez-
Mateo, F.; Calvo-Flores, F. G.; Calvo-Asyn, J. A.; Isac-Garc´ıa, J.; Santoyo-
Gonza´lez, F. Org. Lett. 2003, 5, 1951-1954.
9
J. AM. CHEM. SOC. VOL. 130, NO. 14, 2008 4619

A R T I C L E S
Srinivasachari et al.
Scheme 1. Synthesis of Acetylated Perazido-â-cyclodextrina
a Conditions: (a) PPh3/I2, DMF, 80 °C, 18 h; (b) NaN3, DMF, 80 °C, 24 h; (c) Ac2O, Pyr, 24 h, RT.
Scheme 2. Synthesis of the Alkyne Dendrons (7a-e)a
Scheme 3. Synthesis of the Click Clusters (9a-e)a
a Conditions: (a) CF3COOEt, MeOH; (b) (Boc)2O, CH2Cl2, TEA; (c)
K2CO3, 20:1, MeOH: H2O; (d) propiolic acid, DCC, CH2Cl2.
expression in both HeLa (human cervix carconima) and H9c2
(rat cardiomyoblast) cell lines. In contrast to the positive
controls, these structures reveal low toxicity and do not disrupt
cell surface function in Vitro. The monodisperse macromolecules
designed here represent an incredibly unique multivalent
architecture that can readily be functionalized via noncovalent
(inclusion in the hydrophobic cup) and/or covalent means in a
well-defined manner that can be completely characterized via
conventional spectroscopic methods. This creative motif rep-
resents an important scientific step toward the development of
efficient, nontoxic, fully functionalized, and readily tailorable
macromolecules with discrete molecular weight for a variety
of clinical delivery applications.
Results and Discussion
Synthesis and Characterization of Monomers and Click
Clusters. First, the carbohydrate core, acetylated per-azido-â-
cyclodextrin (4), was synthesized in 72% yield, according to a
combination of previously reported methods (Scheme 1).24,36
Second, we designed a novel series of alkyne dendrons (7a-e)
containing a terminal acetylene group and tert-butoxycarbonyl
(Boc)-protected secondary amines (five structures that varied
in secondary amine stoichiometry, 0-4, Scheme 2). In brief,
the dendrons were synthesized by selectively protecting one of
the terminal primary amine groups of either diethylenetriamine,
triethylenetetramine, tetraethylenepentamine, or pentaethylene-
hexamine (5b-e) with a trifluoroacetyl (COCF3) group, fol-
lowed by in situ protection of the remaining terminal primary
amine and internal secondary amine(s) with (Boc)2O to yield
the Boc derivatives.37 The trifluoroacetyl groups were then
cleaved to yield the Boc-protected monoamines (6b-e, Scheme
a Conditions: (a) CuSO4‚5H2O, sodium ascorbate, tBuOH: H2O (1:1),
70 °C; (b) NaOMe/MeOH, pH ) 9, RT; (c) 4 M HCl/dioxane.
2), which were coupled to propiolic acid via DCC (dicyclo-
hexylcarbodiimide). Purification of the products via silica gel
column chromatography afforded the pure dendrons (7b-e) in
40-60% yield. Dendron 7a was synthesized using a similar
method, by coupling mono-Boc-protected ethylenediamine (6a)
with propiolic acid.
The 1,3-dipolar cycloaddition of azide 4 and each alkyne (7a-
e) was carried out using copper sulfate/sodium ascorbate in 1:1
t-BuOH/H2O to yield the series of protected click clusters (8a-
e, Scheme 3) bearing the regiospecific 1,4-triazole, verified via
NMR, in 80% yield.26,38,39 The products were washed with
(36) Ashton, P. R.; Gattuso, G.; Ko¨niger, R.; Stoddart, J. F.; Williams, D. J. J.
Org. Chem. 1996, 61, 9553-9555.
(37) Geall, A. J.; Taylor, R. J.; Earll, M. E.; Eaton, M. A. W.; Blagbrough, I.
S. Bioconjugate Chem. 2000, 11, 314-326.
(38) Rostovtsev, V. V.; Green, L. G.; Fokin, V. V.; Sharpless, K. B. Angew.
Chem., Int. Ed. 2002, 41, 2596-2599.
9
4620 J. AM. CHEM. SOC. VOL. 130, NO. 14, 2008

Polycationic
â
-Cyclodextrin “Click Clusters”
A R T I C L E S
(Figure 1). It should be noted that this step also aids in the full
removal of copper from the final materials, as chelated copper
may cause unwanted toxicity. The IR spectra also confirmed
the absence of azide and alkyne functionalities. The ESI-MS
showed peaks corresponding only to the fully substituted
materials, as multiply-charged ions (Figure 2). These data
support the formation of pure â-cyclodextrin click clusters (9a-
e) decorated through a 1,2,3-triazole linker with a series of
dendrons containing both internal secondary amines (between
0 and 4) and terminal primary amines that exhibit a cationic
nature at physiological conditions.
Click Cluster-pDNA Binding and Polyplex Size. The final
polycations were studied for their ability to bind and compact
pDNA into nanoparticles using a variety of methods. Prior to
all studies, each cluster was mixed with pDNA at various N/P
ratios (where N ) primary and secondary amines in the click
clusters; P ) phosphate groups in the pDNA) in DNase/RNase-
free water to form the nanoparticle complexes. To confirm
complexation, a gel electrophoresis shift assay was first
performed. Figure 3a shows that compound 9e bound pDNA
at N/P ratios of 1.5 and higher (pDNA lacks migration in the
gel due to polycation binding and charge-neutralization). The
bars in Figure 3c reveal that all of the click clusters fully
complex pDNA between N/P ratios of 1.5 and 2.5, dependent
on the length of the dendron arm. To assess and compare the
relative binding stability, a heparin competitive displacement
assay was performed on the nanoparticles formed with 9a-e
and pDNA at a complexation ratio of N/P ) 5. As pictured in
Figure 3b, complexes formed with 9e and pDNA required a
concentration of 110 µg/mL of heparin to dissociate the binding
(shown by the migration of pDNA in the gel). The binding
affinities of the clusters for pDNA measured via gel shift assay
were corroborated with the heparin displacement results (line,
Figure 3c), where the relative binding stability (heparin con-
centration that dissociated the polyplexes) was found to be 9a
(200 µg/mL) > 9e (110 µg/mL) > 9d (100 µg/mL) > 9c (90
µg/mL) > 9b (80 µg/mL). In general, nanoparticle stability
improved with increase in cluster arm length. This could be
due to an enhancement in electrostatic and H-bonding interac-
Figure 1. 1H NMR of clusters 9a-e in D2O (400 MHz).
ammonium hydroxide to remove trace amounts of copper and
used without further purification. Conventional deprotection of
the acetyl25 and the Boc40 groups afforded the final highly water
soluble click clusters (9a-e) in approximately 50% yield.
Compounds (9a-c) were purified via exhaustive dialysis against
ultrapure water, whereas compounds (9d and 9e) were purified
using cation exchange column with a gradient elution of
ammonium bicarbonate and lyophilized to dryness. The final
compounds 9a-e were characterized using several techniques
to verify complete substitution and pure product formation. The
1
proton peaks were assigned using a combination of H NMR
and 2D COSY experiments. As shown in Figure 1, the 1H NMR
showed the presence of the triazole proton around 8.2 ppm
which indicated the formation of the 1,4-regioisomer exclu-
sively.41 In addition, 2D NOESY (1H-1H) experiments showed
strong NOE effects, due to the close proximity of the triazole
and N-substituted methylene protons, an effect normally ob-
served in the 1,4-regioisomer.39 The absence of peaks at 1.4
ppm (Boc) and 2.0 ppm (acetyl) clearly showed the complete
deprotection of the acetate and carbamate protecting groups
Figure 2. ESI-MS of clusters 9a-e. The labeled masses are for the observed monoisotopic ions. The Mobs is a deconvoluted mass of the compounds from
multiply charged ions.
9
J. AM. CHEM. SOC. VOL. 130, NO. 14, 2008 4621

A R T I C L E S
Srinivasachari et al.
Figure 3. Agarose gel electrophoresis pDNA-click cluster binding assays. (a) Gel shift assay showing 9e-pDNA binding at N/P ratios between 0 (pDNA
only) and 15. Binding is shown by the inhibition of pDNA electrophoretic mobility (band 1). Bands 2 and 3 show the relaxed and supercoiled forms of
pDNA respectively. (b) The concentration of heparin (µgmL-1) required to release pDNA from click cluster-9e binding (polyplexes formed at N/P ) 5).
Heparin dissociation from 9e is shown by the appearance of pDNA electrophoretic mobility. (c) The minimum N/P ratio needed to inhibit electrophoretic
mobility in the click cluster gel shift assays (bars) and the relative binding affinity of each click cluster-pDNA complex as determined by the concentration
of heparin required to release pDNA from each complex (line) at N/P ) 5.
tions as the length of the cluster arms increase (only partial
amine protonation is expected with these clusters, titration
studies are in progress).34,42 We were surprised by the unusually
high heparin concentration needed to dissociate compound 9a,
as it did not follow the general trend, and this result could also
be related to the cluster size. Cluster 9a is the smallest in size
and may partially bind in the pDNA major groove. Therefore,
this complex could be less accessible to competitive displace-
ment by heparin (a large polyanion); if this is the case, smaller
anions could promote more effective displacement (Vide infra).
The compaction of the pDNA into stable nanoparticles in
the size range to be endocytosed (50-500 nm) is an important
process that impacts cellular uptake. Dynamic light scattering
(DLS) and ú potential experiments were conducted to examine
the size and surface charge of the nanoparticles formed at an
N/P ratio of 5. The clusters formed nanoparticles with pDNA
having an average hydrodynamic diameter between 80-130 nm
(Figure 4) that exhibited near neutral to cationic ú potentials.
In general, the ú potential experiments revealed an increase in
positive polyplex potential as the cluster arm length increased.
TEM experiments supported these data, where discrete nano-
particles were observed and imaged (Figure 5).
DNase Protection Assay. The ability of 9a-e to protect
pDNA against nuclease degradation (present in FBS, fetal
bovine serum) was then examined via gel electrophoresis to
verify stable pDNA encapsulation. Nanoparticles were incubated
with FBS at 37 °C for 0, 1, 2, 4, or 8 h. Afterwards, SDS was
added to dissociate the complexes. The ability of each cluster
to protect pDNA from nuclease degradation was determined
by examination of pDNA integrity using electrophoresis. As
shown in Figure 6, the degraded pDNA was visualized by the
appearance of band 6 (degraded small pieces of DNA) and a
Figure 4. The hydrodynamic diameter of the polycation-pDNA complexes
at N/P ) 5 in nuclease free water as determined via dynamic light scattering
(bars) and the ú-potential of the polyplexes (line). All particle size
measurements were found to be statistically different from each other (p <
0.05) except for those indicated with similar geometric red symbols.
decrease in intensity of bands 2-4 (nicked, relaxed, and
supercoiled pDNA, respectively). It should be noted that band
5 arises from a complex formed between FBS and SDS.
Degredation of naked pDNA began immediately upon addition
of FBS, as observed by the decrease in the intensity of bands
2-4, and was fully degraded in only 1 h. Plasmid DNA
complexed with 9b began to degrade after 4 h of incubation
with FBS (shown by the presence of band 6 and the decrease
in intensity of bands 2-4), which is consistent with the gel shift
and heparin displacement assays. This indicates that 9b is the
least effective at binding and protecting pDNA against nuclease
degradation. A slight decrease in intensity of pDNA bands was
observed after incubation of 9a with FBS for 8 h, possibly
indicating some pDNA degradation at this time point and
incomplete pDNA encapsulation. Plasmid DNA complexed with
click clusters 9c, 9d, and 9e was protected from degradation
even after 48 h of exposure to FBS, as shown in Figure 7. These
data indicate that the increased amine stoichiometry within the
dendrons increases pDNA protection from nuclease degradation
and enhances stable encapsulation.
(39) Tornøe, C. W.; Christensen, C.; Meldal, M. J. Org. Chem. 2002, 67, 3057-
3064.
(40) Han, G.; Tamaki, M.; Hruby, V. J. J. Pept. Res. 2001, 58, 338-341.
(41) Fazio, F.; Bryan, M. C.; Blixt, O.; Paulson, J. C.; Wong, C.-H. J. Am.
Chem. Soc. 2002, 124, 14397-14402.
Salt-Induced Polyplex Dissociation. It is important to
understand the unpackaging properties of the click clusters, since
(42) Suh, J.; Paik, H.-J.; Hwang, B. K. Bioorg. Chem. 1994, 22, 318-327.
9
4622 J. AM. CHEM. SOC. VOL. 130, NO. 14, 2008

Polycationic
â
-Cyclodextrin “Click Clusters”
A R T I C L E S
Figure 5. Transmission electron micrograph of the nanoparticles formed with 9a-e and pDNA at N/P ) 5 in water. Nanoparticles were visualized via
negative staining with uranyl acetate, where the bar represents 100 nm.
Figure 6. Protection of pDNA from nuclease degradation (present in FBS)
by 9a (lanes 7-12; gel 1), 9b (lanes 13-18; gel 1), 9c (lanes 1-6; gel 2),
9d (lanes 7-12; gel 2), and 9e (lanes 13-18; gel 2) complexed at N/P )
5. Naked pDNA (lanes 1-6; gel 1) was used as a control. Other control
samples analyzed were (labeled as C for the incubation time): pDNA only
(lane 1, gel 1) and complexes formed with pDNA and 9a (lane 7, gel 1),
9b (lane 13, gel 1), 9c (lane 1, gel 2), 9d (lane 7, gel 2), and 9e (lane 13,
gel 2) without addition of FBS and SDS. Band 1 is the position of the
sample loading (polyplexes without FBS and SDS treatment). Bands 2-4
are nicked, relaxed, and supercoiled pDNA, respectively. Band 5 results
from a complex of FBS and SDS, and band 6 is degraded DNA.
Figure 8. Polyplex dissociation with NaCl(aq). The fraction of dye exclusion
[1 - fraction of PicoGreen intercalation (normalized to PicoGreen intercala-
tion of naked pDNA)] when polyplexes formed with 9a, 9b, 9c, 9d, or 9e
are exposed to increasing concentrations of NaCl.
intercalation into pDNA upon vector release. The concentrations
of NaCl that allowed 50% dye inclusion for click clusters 9a,
9b, 9c, 9d, and 9e were roughly 0.160 M, 0.125 M, 0.280, 0.425,
and 0.440 M, respectively, which represented a qualitative order
of binding strength: 9b < 9a < 9c < 9d < 9e. Therefore, this
study further corroborates the correlation between pDNA
binding and cluster arm length (amine stoichiometry) previously
supported by gel shift, heparin displacement, and nuclease
degradation experiments. In general, as the oligoethyleneamine
length decreased, the pDNA was unpackaged at lower NaCl
concentrations, indicating a weaker binding affinity. Cluster 9b
was found to have the lowest binding strength (unpackaged
pDNA the easiest), which correlates directly to the gel shift
assay, heparin competitive displacements, and nuclease degrada-
tion results that shorter dendrons decrease encapsulation stability.
However, cluster 9a was found to have a binding affinity similar
to (only slightly higher than) 9b, which did not correlate to the
heparin experiments. This experiment supports our hypothesis
that the smallest click cluster, 9a, may partially bind in the
pDNA major groove and could be masked from displacement
with a large polyanion such as heparin. Here, 9a was found to
be readily displaced at relatively low concentrations with a
smaller anion (Cl-), which correlates directly to the nuclease
degradation experiments.
Figure 7. Protection of pDNA from nuclease degradation (present in FBS)
by 9c (lanes 1-4), 9d (lanes 4-8), and 9e (lanes 9-12) complexed at N/P
) 5. Band 1 is the position of the sample loading (polyplexes without FBS
and SDS treatment). Bands 2-4 are nicked, relaxed, and supercoiled pDNA,
respectively. Band 5 results from a complex of FBS and SDS, and band 6
is degraded DNA.
release of therapeutic DNA after its delivery is a crucial process
in transfection. The dissociation of pDNA from the clusters was
studied via PicoGreen, a fluorescent intercalating DNA dye, as
a function of NaCl concentration.43 It is known that increased
salt concentrations can destabilize the binding between nonviral
vectors and DNA by decreasing their electrostatic interaction,
thus leading to decomplexation and/or pDNA release. This
allows increased PicoGreen intercalation, thus resulting in an
enhanced fluorescence. As shown in Figure 8, the PicoGreen
did not intercalate into pDNA molecules complexed by poly-
cations 9a, 9b, 9c, 9d, and 9e, resulting in almost 100%
PicoGreen exclusion, without NaCl addition. However, after the
addition of various concentrations of NaCl (0.1-0.7 M), we
observed an increase in fluorescence due to increased Picogreen
Cellular Delivery and Toxicity Studies. The delivery of
nucleic acids into cells with synthetic materials involves a
complex route that requires the nanoparticles to circumvent
many destructive obstacles during cellular delivery. Subse-
quently, the ability of each click cluster to deliver pDNA into
mammalian cells was screened via flow cytometry experiments
in the HeLa (human cervix adenocarcinoma) and H9c2 (rat
(43) Akinc, A.; Thomas, M.; Klibanov, A. M.; Langer, R. J. Gene. Med. 2005,
7, 657-663.
9
J. AM. CHEM. SOC. VOL. 130, NO. 14, 2008 4623

A R T I C L E S
Srinivasachari et al.
Figure 9. Flow cytometry analysis of pDNA delivery with 9a-e (N/P ) 20) and the controls, DNA only, Jet-PEI (N/P ) 5), and Superfect (N/P ) 5). Both
charts depict the mean fluorescence intensity (bars) and the percentage of cells positive for Cy5 fluorescence (lines) in Opti-MEM conditions for (a) HeLa
cells and (b) H9c2 cells. Each data point represents the mean ( standard deviation of three replicates. All Cy5 fluorescence intensity measurements were
found to be statistically different from each other (p < 0.05) except for those indicated with similar geometric red symbols.
Figure 10. Luciferase gene expression with HeLa and H9c2 cells transfected with each cluster in Opti-MEM. Complexes were formed at N/P ratios of 5,
10, 20, and 50, with each vector (9a-e) and pDNA. The N/P ratio that exhibits the maximum gene expression of the positive controls, Jet-PEI and Superfect,
is 5. (a) Transfection efficiency of vectors in HeLa cells. (b) Transfection efficiency of vectors in H9c2 cells. Each data point represents the mean ( standard
deviation of three replicates. Statistical significance of the data is available in the Supporting Information.
cardiomyoblast) cell lines, in the presence of Opti-MEM (serum-
free media). Jet-PEI and Superfect were used as positive controls
(at N/P ) 5, which is the recommended concentration for most
effective delivery). The flow cytometry experiments were
performed at N/P ) 20, to quantify the cellular uptake of Cy5-
labeled pDNA complexed with 9a-e, and the results are shown
in Figure 9. This N/P ratio was chosen as it represented high
efficacy with low toxicity as determined by gene expression
and toxicity assays (Vide infra). The charts depict the percentage
of cells positive for Cy5-labeled pDNA (lines) and the relative
amount of Cy5-pDNA delivered per cell, displayed as the mean
fluorescence intensity (bars) for HeLa Cells (Figure 9a) and
H9c2 cells (Figure 9b). These data indicate that all of the click
clusters are very effective at promoting cellular uptake of pDNA
in Vitro to HeLa cells. In the HeLa cells, greater than 95% of
cells treated with click cluster-pDNA complexes were positive
for Cy5-labeled pDNA when transfected in Opti-MEM, which
was similar to the positive controls Jet-PEI and Superfect.
However, it should be noted that two cell populations were
observed in the flow cytometry data obtained with Superfect,
which is likely a result of its extreme toxicity in Opti-MEM
(Vide infra). In addition, when comparing the relative mean
pDNA fluorescence intensity (amount of pDNA per cell), the
clusters were about as effective as Superfect; however, Jet-PEI
revealed an order of magnitude lower fluorescence intensity,
indicating that this vector does not carry as much pDNA into
cells as the cluster-based vehicles. In the H9c2 cell line, cellular
uptake of Jet-PEI and Superfect were similar both in the mean
Cy5 fluorescence intensity of cells and percent of cells positive
for Cy5 fluorescence. However, 9a and 9b had significantly
lower cellular uptake then the rest of the vectors. The difference
noticed in the cellular uptake when comparing the HeLa and
H9c2 data may be a result of the difference in the nature between
the cell surfaces. It has previously been demonstrated that the
expression patterns of heparan sulfate proteoglycans varies
drastically in different cell types, with HeLa cells demonstrating
very high expression.44 This increase in anionic proteoglycan
surface environment could permit increased uptake of 9a and
9b in HeLa cells in comparison with H9c2 cells. With H9c2
cells, only vectors 9c-e show high cellular uptake, similar to
that of the positive controls. Altogether, the uptake data for both
cell lines indicated that the click clusters with the longest
oligoethyleneamine arms are most effective at delivering pDNA
into differing cell types in Opti-MEM.
The transfection efficiency of the click clusters were also
studied by performing luciferase reporter gene expression
experiments with HeLa and H9c2 cells at a variety of N/P ratios
(5, 10, 20, and 50), in Opti-MEM. Jet-PEI and Superfect were
again used as positive controls. The luciferase gene expression
profiles for both cell lines, shown in Figure 10, indicate that
the delivery efficiency generally increased with the N/P ratio
for clusters 9a-e. With HeLa cells, 9d had the highest
transfection efficiency of all the click clusters, whereas 9a and
9b showed only moderate efficacy even at high N/P ratios. It
(44) Pajusola, K.; Gruchala, M.; Joch, H.; Lu¨scher, T. F.; Yla¨-Herttuala, S. J.
Virol. 2002, 76, 11530-11540.
9
4624 J. AM. CHEM. SOC. VOL. 130, NO. 14, 2008

Polycationic
â
-Cyclodextrin “Click Clusters”
A R T I C L E S
Figure 11. Cell viability of (a) HeLa cells and (b) H9c2 cells 48 h after polyplex exposure as measured by protein content. Polyplexes were formed with
9a-e and pDNA at N/P of 5, 10, 20, and 50, and incubated in Opti-MEM. The polyplexes were formulated Jet-PEI and Superfect and N/P ) 5. Each data
point represents the mean ( standard deviation of three replicates. Statistical significance of the data is available in the Supporting Information.
transfections of both cell lines revealed that 9d and 9e are
similarly as efficient as the positive controls. The high efficacy
of these two click clusters is likely due to a combination of
features. Vectors 9d and 9e deliver pDNA into 95-100% of
HeLa and H9C2 cells, and these two vectors exhibit high
stability from salt and heparin-induced unpackaging. Also, 9d
and 9e prohibited pDNA denegration, and therefore likely
promotes a higher delivery rate of expression-competent pDNA
during the transfection process. It should be noted that when
the data were normalized to the milligrams of protein, Jet-PEI
appears artificially high due to the high toxicity (very low
amount of protein present) (see Supporting Information). With
HeLa cells, Superfect was highly toxic and the protein content
was virtually undetectable (thus transfection efficiency was
unable to be normalized to milligrams of protein; see Supporting
Information).
The cell viability profiles were first measured via protein
content in both the HeLa and H9c2 cell lines after transfection
in Opti-MEM (Figure 11). The results demonstrate that the click
clusters have minimal cytotoxic effects to cultured cells (viability
at or above 75%) even at an N/P of 50. Jet-PEI and Superfect
were found to exhibit high toxicity with both cell lines when
compared to click clusters. It should be noted here that Jet-PEI
is a polydisperse material and Superfect is a fractured dendrimer.
Therefore, in addition to the toxicity, the synthesis of identical
batches of Jet-PEI and Superfect could be an issue at a clinical
scale. The above promising results show that these discrete
Figure 12. (a) MTT assay of 9a-e in HeLa cells after 32 h of exposure
monodisperse cyclodextrin-based click clusters have high bio-
compatibility and efficacy in Vitro and warrant investigation
for in ViVo nucleic acid delivery, which is planned in our future
studies.
to vector. Cells were treated with the click clusters at four different
concentrations correlating to the N/P ratios of 150, 50, 20, and 5 used in
transfection studies. Jet-PEI and Superfect vectors were administered to
cells at six concentrations similar to those used to formulate polyplexes at
N/P ) 5. Each data point represents the mean ( standard deviation of three
separate measurements. (b) Statistical analysis of the cell viability of HeLa
cells measured via MTT assay. Bars indicate the percentage of cells viable
(at N/P 5 for Jet-PEI and Superfect; at N/P 20 for the click clusters). All
measurements were found to be statistically different from each other (p <
0.05) except for those indicated with similar geometric red symbols.
An MTT assay was performed on HeLa and H9c2 cells
treated with the click clusters (not polyplex) to get a more
accurate assessment of the cytotoxicity (Figure 12). This assay
relies on the rate of a crucial mitochondrial enzyme as a measure
of cell survival. In this assay, the results correlate with the
protein assay (Figure 11). Clusters 9a and 9b appear to decrease
cell viability at very high vector concentrations. This experiment
further demonstrates the toxicity of the positive controls,
revealing that concentrations less than 0.1 µg/mL cause complete
cell death. The positive controls, Superfect and Jet-PEI,
exhibited an LD50 value of 0.0073 µg/uL and 0.0010 µg/uL,
respectively. These concentrations are much less than those used
when formulating polyplexes at an N/P ratio of 5. An LD50 value
was not a surprise to find that 9b displayed the lowest gene
expression, which was consistent with the previous data and
likely due to the weak binding affinity and lack of pDNA
protection against enzymatic degradation. Transfection in the
H9c2 cell line supports that 9d and 9e are most efficacious click
clusters. Although the trends differ slightly between cell lines,
differences in growth rate and intracellular trafficking mecha-
nisms may be the cause for the observed differences. However,
9
J. AM. CHEM. SOC. VOL. 130, NO. 14, 2008 4625

A R T I C L E S
Srinivasachari et al.
Figure 13. Membrane integrity of HeLa cells reported as the percentage of lactate dehydrogenase (LDH) release 5 h after exposure to (a) polyplexes formed
with Jet-PEI, Superfect, 9a, 9b, 9c, 9d, or 9e and (b) Jet-PEI, Superfect, 9a, 9b, 9c, 9d, or 9e only (no pDNA). An LDH assay kit was used to measure the
amount of LDH enzyme leakage into the culture the media. A control of lysed cells was used to determine the total amount of LDH present in each sample
and normalize to 100% LDH released. Each data point represents the mean ( standard deviation of three separate measurements.
could only be calculated for two clusters, 9a (0.3525 µg/uL)
and 9b (0.5073 µg/uL). Both of these concentrations are higher
than the concentrations needed to form polyplexes at an N/P
ratio of 50. The toxicity of the other click clusters was not high
enough to measure LD50 values in the concentration range of
this experiment. The MTT assays with H9c2 cells showed very
similar results (data not shown).
membrane. Increasing the amount of Superfect does not appear
to further increase damage. Conversely, Jet-PEI alone exhibits
some membrane damage at low N/P ratios, but the amount of
LDH released increases drastically with the vector dosage, where
nearly 80% of LDH release was found. Conversly, the click
clusters did not exhibit any evidence of membrane damage, even
when cells were treated with vector concentrations much higher
than Jet-PEI and Superfect. Therefore, the â-cyclodextrin cups
present in the click clusters are not capable of inducing
membrane damage. However, increased cell surface binding via
â-cyclodextrin inclusion of cell surface groups may increase
cellular uptake. This hypothesis is supported by cellular uptake
data (Figure 9a) which demonstrates that even vectors with short
oligoethyleneamine arms (9a, 9b) elicit high cellular uptake of
pDNA.
In order to further investigate the biological impact of these
polycationic click clusters on cells, an LDH (lactate dehydro-
genase) assay was used to measure the membrane integrity of
cells treated both with vector alone and with cluster-pDNA
nanoparticles. It has previously been demonstrated that very
charge dense cations such as PEI and Superfect can damage
the plasma membrane of cells, which contributes to the
cytotoxicity of these vectors.45 Furthermore, we wanted to
investigate the effect of the â-cyclodextrin-based macromol-
ecules on membrane integrity. It is well-known that a cyclo-
dextrin cup can complex cholesterol and other biological
molecules (present on the cell surface); that this may lead to
membrane damage and subsequent leakage of LDH from the
cytosol. This assay measures the amount of LDH released into
the media and uses it as an indication of cell membrane
disruption. Controls of lysed cells were used to determine the
total amount of LDH present in each sample. This value was
used to normalize the experimental samples and determine the
total amount of LDH released from these cells (Figure 13).
When cells were treated with polyplexes (Figure 13a), it was
found that Jet-PEI and Superfect polyplexes do not damage the
cell membrane at low N/P ratios. However, when the polyplex
N/P ratios were increased, significant amounts of LDH were
found in the sample media indicating that these vectors induced
membrane damage. When this assay was performed with
polyplexes formed with the click clusters, full membrane
integrity was observed at all N/P ratios. When cells were treated
with the vectors alone (at the respective concentration used to
formulate the polyplexes show in Figure 13a), Superfect was
found to release 65-70% of cytosolic LDH (Figure 13b),
indicating that this vector is very damaging to the cell
Conclusion
In summary, a novel series of polycationic â-cyclodextrin
“click clusters” have been synthesized by linking a per-azido-
â-cyclodextrin core moiety to oligoethyleneamine dendrons via
click coupling chemistry. The final clusters, 9a-e, were found
to be completely substituted and deprotected according to a
variety of characterization techniques. These discrete macro-
molecules are highly water soluble, and most of the click clusters
were found to complex and protect pDNA in a stable manner
with size and morphologies similar to viral delivery vehicles.
Cellular delivery and viability experiments demonstrate that all
of these structures have the ability to deliver Cy5-labeled pDNA
and pDNA encoding the luciferase reporter gene into HeLa and
H9c2 cells very effectively in Opti-MEM, with low cytotoxic
effects. In particular, 9d and 9e, with the longest dentron arms,
were the most effective delivery vehicles.
The synthetic design and application of these well-defined
and characterized structures is very significant and relevant to
the development of clinically viable drug delivery vehicles. For
example, the terminal primary amines can be symmetrically
functionalized with various biologically active groups (such as
drug molecules or imaging agents). The hydrophobic cup can
(45) Moghimi, S. M.; Symonds, P.; Murray, J. C.; Hunter, A. C.; Debska, G.;
Szewczyk, A. Mol. Ther 2005, 11, 990-995.
9
4626 J. AM. CHEM. SOC. VOL. 130, NO. 14, 2008

Polycationic
â
-Cyclodextrin “Click Clusters”
A R T I C L E S
be exploited to host adamantane groups substituted with
poly(ethylene glycol) chains to increase stability from aggrega-
tion, circulation lifetime, and/or tissue-specific targeting of the
nanoparticles for enhanced therapeutic selectivity and specific-
ity.31 Future experiments are aimed at exploring these unique
macromolecules for a variety of biomedical applications.
was supported by the NSF CAREER (CHE-0449774) and the
Beckman Young Investigators Programs.
Supporting Information Available: Experimental details of
the synthesis and characterization of compounds 4, 7a -e, and
9a -e and details of the electrophoresis, ethidium bromide
exclusion assays, TEM, dynamic light scattering, cell culture,
flow cytometry, reporter gene, and viability assays. This material
is available free of charge via the Internet at http://pubs.acs.org.
Acknowledgment. The authors gratefully acknowledge Dr.
Larry Sallans (Mass Spec Facility, Dept. of Chemistry, Uni-
versity of Cincinnati) for his immense help in the ESI-MS
analyses of the compounds reported in this paper. This work
JA074597V
9
J. AM. CHEM. SOC. VOL. 130, NO. 14, 2008 4627