Efficient nucleic acid transduction with lipoplexes containing novel piperazine- and polyamine-conjugated cholesterol derivatives

Article abstract of DOI:10.1016/j.bmcl.2008.11.009

To advance the use of cationic lipids for non-viral nucleic acid vector formulation, a panel of novel nitrogen heterocycle cholesteryl derivatives containing a biodegradable carbamate linker was synthesised. Optimally acting piperazine and cyclen compound

Full text of DOI:10.1016/j.bmcl.2008.11.009

Bioorganic & Medicinal Chemistry Letters 19 (2009) 100–103
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Efficient nucleic acid transduction with lipoplexes containing novel piperazine-
and polyamine-conjugated cholesterol derivatives
Rafique Ul Islam a,b, , Justin Hean a, , Willem A. L. van Otterlo b, Charles B. de Koning b, Patrick Arbuthnot a,
*
a Antiviral Gene Therapy Research Unit, Department of Molecular Medicine and Haematology, University of the Witwatersrand Medical School, Private Bag 3, WITS 2050, South Africa
b Molecular Sciences Institute, School of Chemistry, University of the Witwatersrand, South Africa
a r t i c l e i n f o
a b s t r a c t
Article history:
To advance the use of cationic lipids for non-viral nucleic acid vector formulation, a panel of novel nitro-
gen heterocycle cholesteryl derivatives containing a biodegradable carbamate linker was synthesised.
Optimally acting piperazine and cyclen compounds had nucleic acid-binding and lipoplex nanoparticle
formation properties that were suitable for their use as non-viral vectors. It was found that the lipoplexes
formed were capable of efficient non-toxic nucleic acid delivery to cells in culture. The chemical structure
of individual cationic lipids, which is likely to influence lipoplex formation, affected efficiency of DNA or
RNA transfection. The results indicated that the cyclen containing compound possessing two cholesteryl
moieties resulted in efficient siRNA-mediated target gene silencing but was a poor reagent for DNA
transfection.
Received 25 September 2008
Revised 31 October 2008
Accepted 3 November 2008
Available online 6 November 2008
Keywords:
Lipoplex
Cholesterol
Cyclen
Ó 2008 Elsevier Ltd. All rights reserved.
DNA
RNA
Transfection
Piperazine
Gene therapy has enormous potential for the management of
many serious diseases. In addition to supplementing cellular ge-
netic material, use of nucleic acids to activate RNA interference
(RNAi) and silence pathology-causing gene expression has prom-
ising application.1 However, achieving efficient and safe delivery
of nucleic acids remains an important objective before the full
potential of this approach is realized. Viral vectors have been
extensively used, but complications of toxicity, mutagenesis
and immunostimulation are problematic.2 Non-viral vectors are
usually chemically synthesized and have advantages of potential
for large scale production and modification to alter biological
properties. Nevertheless, improvement of nucleic acid delivery
efficiency is important to achieve before widespread application
of non-viral vectors is realized. DNA and RNA complexed to lip-
osomes to form lipoplexes are commonly used for transfection of
cells in culture and nucleic acid delivery in vivo.3 Lipoplexes
typically comprise nucleic acids bound to cationic constituents
together with a helper/fusogenic lipid. Incorporation of cationic
nucleic acid-binding cholesterol-based lipids into non-viral vec-
tors has gained favor and this class of cationic lipid has been
extensively studied.4–11 Cationic cholesterol derivatives have
three basic domains: a cholesteryl (hydrophobic) part, a cationic
amine head group and a linker moiety.8,10 Incorporation of a
variety of polar head groups has been reported to bind DNA
and achieve nucleic acid delivery to cells. These include linear
polyamines such as spermine and spermidine,12 dendritic
amines,13 piperazines,14 heterocyclic and aromatic heterocyclic
substituted amines.15 Chemical properties of individual constitu-
ents as well as their formulation with other components of the
lipoplexes influence transduction efficiency. To advance the
favorable gene delivery properties of piperazine-containing for-
mulations,14 we have carried out analysis on a range of novel
piperazine cholesteryl compounds and also extended the study
to include carbamate-linked polyamine (cyclen) cholesterol
derivatives. We have found that lipoplexes containing these
lipids are efficient and non-toxic vectors for delivery of both
DNA and synthetic RNAi effecters to cells in culture.
Substituted piperazine/cyclic polyamine (cyclen) derivatives
were synthesized by reacting cholesteryl chloroformate with cyclic
nitrogen containing compounds using a convenient acylation pro-
cedure (Fig. 1). Some of the compounds were not subjected to de-
tailed analysis because of poor solubility or low DNA binding
capacity. Briefly, synthesis reaction conditions were as follows.
Cholesteryl chloroformate (1.0 mmol) in dichloromethane
(10 cm3) was added slowly (10 min) to a solution of amines/poly-
amines (1.2 mmol) in dichloromethane (10 cm3) at 0 °C. The reac-
tion solution was stirred for 10 min at 0 °C and then for a further
4–8 h at room temperature. Thereafter, the reaction mixture was
washed with water (3 ꢀ 10 cm3) and dried (MgSO4) which afforded
the product as a solid. The crude solid was subsequently purified
* Corresponding author. Tel.: +27 (0)11 717 2365; fax: +27 (0)11 717 2395.
E-mail address: Patrick.Arbuthnot@wits.ac.za (P. Arbuthnot).
These authors contributed equally to the work.
0960-894X/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2008.11.009

R. U. Islam et al. / Bioorg. Med. Chem. Lett. 19 (2009) 100–103
101
Figure 1. Cationic cholesterol derivatives 1–19 used in this study. (A) Schematic representation of the chemical synthesis. (B) Structures of moieties (X) attached to
cholesterol derivatives 1–19.
by silica gel column chromatography using 2–10% methanol-
dichloromethane as eluant. This gave the desired conjugated com-
pounds 1–19 in good to excellent yields. All compounds were char-
acterized by NMR, IR and mass spectroscopy and gave acceptable
Figure 2. Representative analysis of properties of lipoplexes formulated with individual piperazine cholesteryl derivatives. (A) Mobility shift assay to determine DNA binding
by liposomes. Varying ratios of liposome:DNA ( g: g) were combined then subjected to agarose gel electrophoresis. Bound uncharged nucleic acid was retarded in the wells,
l
l
while uncomplexed DNA retained electrophoretic mobility. (B) Nuclease protection of liposomes was determined by incubating naked and complexed DNA in the presence of
DNase-containing serum. This was followed by electrophoresis to assess evidence for nucleic acid degradation. Lipoplex particle size analysis of formulations containing
cationic cholesterol derivative 4 (C) and 5 (D). Malvern particle size analysis was carried out to determine the lipoplex particle sizes at varying liposome: DNA (lg:lg) ratios.

102
R. U. Islam et al. / Bioorg. Med. Chem. Lett. 19 (2009) 100–103
data. Cationic lipid 5 was formulated with dioleylphosphatidyl eth-
anolamine (DOPE) at a 7.2:4 molar ratio, while all of the other lip-
ids were mixed at a ratio of 6:4. Formation of complexes with DNA
was carried out using standard methodology.16
Initially, mobility shift analysis was carried out to determine
the DNA binding capabilities of the soluble cholesteryl amine-con-
taining derivatives, (Fig. 2A). Substituted piperazine-containing
aromatic (10) and heteroaromatic (6,7) rings were found to be
insoluble. DNA was incubated at varying ratios (lg:lg) to cationic
lipid prior to electrophoresis. The ratio at which retardation of the
DNA occurred was used as an indicator of the point at which neu-
tralization of the DNA charge was achieved. This charge neutraliz-
ing ratio varied with different compounds and ranged from 3:1
(e.g., lipid 10) to 9:1 (e.g., lipid 4) for cationic lipids with decreasing
DNA binding capabilities (Table 1). The DC-Chol (lipid 3) standard
was capable of neutralizing DNA charge at a ratio of 5:1, which is
similar to previously reported DNA binding capacity.5 Stability of
the DNA within the particle formulations was assessed after expo-
sure to serum containing DNases and representative analysis is
shown in Figure 2B. Our findings were that formulation of the
DNA within the lipoplexes was protected against degradation by
nucleases, which is important for use of nucleic acid complexes
for gene delivery. To characterize the physical properties of the lip-
oplex nanoparticles further, formulations were subjected to Mal-
vern particle size analysis. Formulations were prepared without
extrusion and under conditions where DNA charge was neutralized
at minimum lipid concentration. The average particle diameter
ranged from 150 to 300 nm (Table 1). Sizes also varied according
to the amount of lipid incorporated into the complexes (Fig. 2C
and D). With formulations containing lipid 4, increasing the ratio
of lipid:DNA from 0:1 to 10:1 resulted in a steady increase of lipo-
plex diameter from 280 to 750 nm. Conversely, increasing the con-
tent of lipid 5 within the lipoplex resulted in an increase of particle
diameter size to 350 nm at a ratio of 5:1 and thereafter, the particle
size diameter diminished to 150 nm with lipid:DNA ratios above
7:1. These particle sizes are predicted to be suitable for use
in vivo, where diameters of below 150 nm are needed to penetrate
certain membrane fenestrations and to avoid consumption by the
reticuloendothelial system.10
Figure 3. Analysis of nucleic acid transfection efficiency and toxicity of individual
lipoplexes. (A) Comparison of luciferase reporter gene activity following transfec-
tion of Huh-7 cells with lipoplexes formulated with each of the indicated piperazine
cholesterol derivatives. Representative fluorescence microscopy fields to detect
eGFP-expressing Huh-7 cells following transfection with lipoplexes containing DC-
Chol or cationic lipid 5 are shown in the inset. (B) Inhibition of HBV surface antigen
secretion from Huh-7 cells following siRNA transfection. Cells were initially
transfected with a replication-competent HBV plasmid and treated with the siRNA
formulations 24 h thereafter. Controls included naked siRNA and a commercially
available transfection reagent (Lipofectamine 2000). (C) Toxicity analysis of
lipoplexes containing DC-Chol, cationic lipids 5 and 6 which were formulated at
varying ratios of DNA to liposome.
To determine DNA transfection efficiency of lipoplexes, en-
hanced green fluorescent protein (eGFP) and luciferase reporter gene
transfection to cultured cells was used. Comparison revealed that
incorporation of cationic lipid 4 into lipoplex compounds achieved
the best transfection efficiency, which was also similar to that of
DC-Chol (Fig. 3A and B). Interestingly, considerable variation was
observed with formulations that contained different cholesteryl
piperazine and cyclen compounds. To characterize RNA transfec-
tion properties of lipoplexes, formulations were prepared that con-
tained anti-hepatitis B virus (HBV) siRNA. The siRNA duplex
comprised 50 UUGAAGUAUGCCUCAAGGUCG 30 antisense and 50
ACCUUGAGGCAUACUUCAATT 30 sense strands, which targeted
HBV coordinates 1700–1720 (Genbank Accession No. J02203). Cells
were initially transfected with a HBV replication-competent plas-
mid17 and thereafter with formulations containing anti HBV siRNA.
Knockdown of HBV surface antigen (HBsAg), a marker of viral rep-
lication in culture, was determined using ELISA assay.18 Compared
to mock transfected cells, the siRNA delivered within lipoplex 5
achieved significant inhibition of secretion of HBsAg from the
transfected cultured cells (Fig. 3C). Silencing efficiency was similar
to that achieved with a commercially available transfection re-
agent (Lipofectamine 2000).
To assess unintended off target effects of the formulations, tox-
icity was assessed using the Cytotox Glo kit (Promega, WI, USA).
Cell injury induces release of proteases that cleave 3 amino acids
from a luciferin precursor to generate a mature luciferase substrate
and emitted light is used as an index of cell toxicity. Under stan-
dard conditions of transfection, formulations were essentially
non-toxic, but with addition of more lipids, toxicity increased. Rep-
resentative toxicity assessment of formulations containing lipids 4
and 5 was similar to that observed for the control DC-Chol formu-
lations (Fig. 3D). Induction of a non-specific interferon response, as
assessed by activation of interferon-b gene expression, was also
determined. We observed that transfected siRNA did not induce
activation of this gene (not shown).
Table 1
Minimum liposome to DNA ratio, with particle diameters for selected lipoplexes, at
which nucleic acid charge neutralization occurs within the formulations.
Cationic lipid
Optimal DNA
binding ratio (
Lipoplex particle
diameter (nm)
lg:lg)
1
2
No binding
5:1
3
5:1
240
The nucleic acid-binding lipid components of lipoplex vectors
have an influence on their transfection efficiency and the overall
behavior of liposome-derived non-viral vectors. Identification of
new cationic compounds and optimization of formulations for
use in therapy is therefore an important objective to advance
4
5
8
9
5:1
5:1
4:1
7:1
158.8
524.2
10
3:1
766.8

R. U. Islam et al. / Bioorg. Med. Chem. Lett. 19 (2009) 100–103
103
therapeutic nucleic acid transfer. In this report, we have shown
that cationic cholesteryl piperazines and cyclen derivatives, which
were synthesized using uncomplicated protocols, can be conve-
niently formulated in lipoplex non-viral particles. Properties of
DNA binding, nanoparticle size, protection from nuclease degrada-
tion and efficiency of nucleic acid delivery were found to be useful
for achieving efficient nucleic acid transfer. Vectors containing the
cationic lipids functioned well as transfecting reagents and were
capable of delivery of either plasmid DNA or siRNA. Efficiency
was achieved which is similar to that of standard and commer-
cially available reagents, however the transfection efficiency of for-
mulations differed according to the type of cationic cholesterol
derivative contained within the vector. This observation is likely
to be a result of particular properties of the 2 types of nucleic acid
tested. Although both DNA and RNA have the same anionic charge
to mass ratio, their behavior within lipoplexes is different. Neutral-
ization of anionic phosphates of circular duplex plasmid DNA leads
to controlled condensation during the formation of the lipoplex.
However, because of their small size, siRNA interaction with cat-
ionic lipids may be different and lead to partial encapsulation,
irregularly sized particles and poor stability of lipoplexes. Our
observations indicate that cationic compound 4 favors an ordered
condensation of plasmid DNA, which does not occur with lipid 5.
Conversely, siRNA lipoplexes containing compound 5 are capable
of good RNA transfection efficiency. This interesting distinction is
likely to result from differing interaction between plasmid DNA
or siRNA with each of the cationic cholesteryl piperazines. Cationic
lipid 5 contains two cholesteryl moieties, which are covalently
linked by a aza-macrocycle.19 However, lipid 4 has a more conven-
tional structure with single cholesterol attached to DNA binding
cationic head group. Dual cholesterol moieties within the lipoplex
may favor controlled interaction of the smaller siRNAs to form a
lipoplex that is capable of efficient delivery of the siRNAs to their
cytoplasmic site of action. The promising properties of the trans-
fection reagents which we have observed have prompted us to ini-
tiate assessment of their use for nucleic acid delivery in vivo.
Acknowledgments
This work was supported by funding under the Sixth Research
Framework Programme of the European Union, Project RIGHT
(LSHB-CT-2004-005276), from CANSA, the South African National
Research Foundation (NRF, GUN 68339, 65495, 2053652 and RNA
‘Synthesis and biological screening’), ESASTAP and the South Afri-
can Poliomyelitis Research Foundation.
References and notes
1. Dykxhoorn, D. M.; Palliser, D.; Lieberman, J. Gene Ther. 2006, 13, 541.
2. Raty, J. K.; Lesch, H. P.; Wirth, T.; Yla-Herttuala, S. Curr. Drug Saf. 2008, 3, 46.
3. Kostarelos, K.; Miller, A. D. Adv. Genet. 2005, 53PA, 69.
4. Belmont, P.; Aissaoui, A.; Hauchecorne, M.; Oudrhiri, N.; Petit, L.; Vigneron, J. P.;
Lehn, J. M.; Lehn, P. J. Gene Med. 2002, 4, 517.
5. Gao, X.; Huang, L. Biochem. Biophys. Res. Commun. 1991, 179, 280.
6. Hasegawa, S.; Hirashima, N.; Nakanishi, M. Bioorg. Med. Chem. Lett. 2002, 12,
1299.
7. Hattori, Y.; Ding, W. X.; Maitani, Y. J. Control Release 2007, 120, 122.
8. Karmali, P. P.; Chaudhuri, A. Med. Res. Rev. 2007, 27, 696.
9. Lee, Y.; Koo, H.; Lim, Y. B.; Lee, Y.; Mo, H.; Park, J. S. Bioorg. Med. Chem. Lett.
2004, 14, 2637.
10. Martin, B.; Sainlos, M.; Aissaoui, A.; Oudrhiri, N.; Hauchecorne, M.; Vigneron, J.
P.; Lehn, J. M.; Lehn, P. Curr. Pharm. Des. 2005, 11, 375.
11. Singh, R. S.; Chaudhuri, A. FEBS Lett. 2004, 556, 86.
12. Blagbrough, I. S.; Geall, A. J.; Neal, A. P. Biochem. Soc. Trans. 2003, 31, 397.
13. Haensler, J.; Szoka, F. C., Jr. Bioconjug. Chem. 1993, 4, 372.
14. Gao, H.; Hui, K. M. Gene Ther. 2001, 8, 855.
15. Ilies, M. A.; Seitz, W. A.; Caproiu, M. T.; Wentz, M.; Garfield, R. E.; Balaban, A. T.
Eur. J. Org. Chem. 2003, 2645.
16. Singh, M.; Kisoon, N.; Ariatti, M. Drug Deliv. 2001, 8, 29.
17. Nassal, M. J. Virol. 1992, 66, 4107.
18. Carmona, S.; Ely, A.; Crowther, C.; Moolla, N.; Salazar, F. H.; Marion, P. L.; Ferry,
N.; Weinberg, M. S.; Arbuthnot, P. Mol. Ther. 2006, 13, 411.
19. Venkata Ramana, A.; Watkinson, M.; Todd, M. H. Bioorg. Med. Chem. Lett. 2008,
18, 3007.